Skip to main content

Full text of "Scientific misconduct in breast cancer research : hearings before the Subcommittee on Oversight and Investigations of the Committee on Energy and Commerce, House of Representatives, One Hundred Third Congress, second session, April 13 and June 15, 1994"

See other formats


' SCIENTinC MISCONDUCT IN BREAST CANCER 

RESEARCH 

Y 4, EN 2/3: 103-143 ^~ 

Scientific llisconduct in Breast Can... 

HEARINGS 

BEFORE THE 

SUBCOMMITTEE ON 
OVERSIGHT AND INVESTIGATIONS 

OF THE 

COMMITTEE ON 

ENERGY AND COMMERCE 

HOUSE OF REPRESENTATIVES 

ONE HUNDRED THIRD CONGRESS 

SECOND SESSION 



APRIL 13 and JUNE 15, 1994 



Serial No. 103-143 



Printed for the use of the Committee on Energy and Commerce 




r 

--* .I-— ■-- :, - 



'^^ 8 re. 



U.S. GOVERNMENT PRINTING OFFICE 
84-5 lOCC WASHINGTON : 1994 

For sale by the U.S. Government Printing Office 
Superintendent of Documents, Congressional Sales Office, Washington, DC 20402 
ISBN 0-16-046276-2 



' SCIENTinC MISCONDUCT IN BREAST CANCER 

RESEARCH 

Y 4. EN 2/3: 103-143 ^~ 

Scientific Hisconduct in Breast Can... 

HEARINGS 

BEFORE THE 

SUBCOMMITTEE ON 
OVERSIGHT AND INVESTIGATIONS 

OF THE 

COMMITTEE ON 

ENERGY AND COMMERCE 

HOUSE OF REPRESENTATIVES 

ONE HUNDRED THIRD CONGRESS 

SECOND SESSION 



APRIL 13 and JUNE 15, 1994 



Serial No. 103-143 



Printed for the use of the Committee on Energy and Commerce 




/ 



- - 4 dy 



4^. 



U.S. GOVERNMENT PRINTING OFFICE 



84-5 lOCC WASHINGTON : 1994 



For sale by the U.S. Government Printing Office 
Superintendent of Documents, Congressional Sales Office, Washington, DC 20402 
ISBN 0-16-046276-2 



COMMITTEE ON ENERGY AND COMMERCE 



JOHN D. DINGELL, Michigan, Chairman 



HENRY A. WAXMAN, CaUfornia 

PHILIP R. SHARP, Indiana 

EDWARD J. MARKEY, Massachusetts 

AL SWIFT, Washington 

CARDISS COLLINS, IlUnois 

MIKE SYNAR, Oklahoma 

W.J. "BILLY" TAUZIN, Louisiana 

RON WYDEN, Oregon 

RALPH M. HALL, Texas 

BILL RICHARDSON, New Mexico 

JIM SLATTERY, Kansas 

JOHN BRYANT, Texas 

RICK BOUCHER, Virginia 

JIM COOPER, Tennessee 

J. ROY ROWLAND, Georgia 

THOMAS J. MANTON, New York 

EDOLPHUS TOWNS, New York 

GERRY E. STUDDS, Massachusetts 

RICHARD H. LEHMAN, California 

FRANK PALLONE, Jr., New Jersey 

CRAIG A. WASHINGTON, Texas 

LYNN SCHENK, California 

SHERROD BROWN, Ohio 

MIKE KREIDLER, Washington 

MARJORIE MARGOLIES-MEZVINSKY, 

Pennsylvania 
BLANCHE M. LAMBERT, Arkansas 

Alan J. Roth, Staff Director and Chief Counsel 

Dennis B. Fitzgibbons, Deputy Staff Director 

Margaret A. Durbin, Minority Chief Counsel and Staff Director 



CARLOS J. MOORHEAD, CaUfornia 

THOMAS J. BLILEY, Jr., Virginia 

JACK FIELDS, Texas 

MICHAEL G. OXLEY, Ohio 

MICHAEL BILIRAKIS, Florida 

DAN SCHAEFER, Colorado 

JOE BARTON, Texas 

ALEX MCMILLAN, North Carolina 

J. DENNIS HASTERT, IlUnois 

FRED UPTON, Michigan 

CLIFF STEARNS, Florida 

BILL PAXON, New York 

PAUL E. GILLMOR, Ohio 

SCOTT KLUG, Wisconsin 

GARY A. FRANKS, Connecticut 

JAMES C. GREENWOOD, Pennsylvania 

MICHAEL D. CRAPO, Idaho 



Subcommittee on Oversight and Investigations 



JOHN D 

SHERROD BROWN, Ohio 

MARJORIE MARGOLIES-MEZVINSKY, 

Pennsylvania 
HENRY A. WAXMAN, CaUfornia 
CARDISS COLLINS, Illinois 
RON WYDEN, Oregon 
JOHN BRYANT, Texas 

Reid P.F. 
Peter 



DINGELL, Michigan, Chairman 

DAr«I SCHAEFER, Colorado 
CARLOS J. MOORHEAD, CaUfornia 
JOE BARTON, Texas 
FRED UPTON, Michigan 



Stuntz, Staff Director I Chief Counsel 
D.H. Stockton, Research Analyst 
Bruce F. Chafin, Special Assistant 
John J. Hambel, Minority Counsel 



(II) 



CONTENTS 



Page 

Hearings held on: 

April 13, 1994 1 

June 15, 1994 101 

Testimony of: 

Bivens, Lyle W., Director, Office of Research Integrity, Department of 
Health and Human Services 51 

Broder, Samuel, Director, National Cancer Institute 45, 198 

Chabner, Bruce, Director, Division of Cancer Treatment, National Cancer 

Institute 45 

Detre, Thomas, senior vice chancellor of health sciences. University of 
Pittsburgh 134 

Fisher, Bernard, former chairman. National Surgical Adjuvant Breast 
and Bowel Project, University of Pittsburgh 168 

Friedman, Michael A., Associate Director, Cancer Therapy Evaluation 

Program, National Cancer Institute 45 

Herberman, Ronald B., interim chairman. National Surgical Adjuvant 
Breast and Bowel Project, University of Pittsburgh 138 

Milbauer, Alan, vice president, external affairs, Zeneca Pharmaceuticals 
Group 107 

O'Connor, J. Dennis, chancellor, University of Pittsburgh 132 

Onex, Joseph, counsel to Bernard Fisher, University of Pittsburgh 168 

Patterson, John, international medical director, Zeneca Pharmaceuticals 
Group 107 

Pearson, Cjrnthia A., program director, National Women's Health Net- 
work 20 

Plourde, Paul, senior director, Clinical and Medical Affairs, Zeneca Phar- 
maceuticals Group 107 

Schroeder, Hon. Patricia, a Representative in Congress from the State 
of Colorado 8 

Sigal, Jill Lea, consultant, Washington, DC 28 

Snowe, Hon. Olympia, a Representative in Congress from the State of 
Maine 10 

Varmus, Harold, Director, National Institutes of Health 43 

Visco, Fran, president, National Breast Cancer Coalition 12 

(III) 



SCIENTIFIC MISCONDUCT IN BREAST 
CANCER RESEARCH 



WEDNESDAY, APRIL 13, 1994 

House of Representatives 
Committee on Energy and Commerce, 
Subcommittee on Oversight and Investigations 

Washington, DC. 

The subcommittee met, pursuant to notice, at 10:05 a.m., in room 
2123, Rayburn House Office Building, Hon. John D. Dingell (chair- 
man) presiding. 

Mr. Dingell. The subcommittee will come to order. 

Today the subcommittee will examine a number of important is- 
sues associated with serious falsification and fabrication of data in 
some of the Nation's most important clinical trials on the treat- 
ment and prevention of breast cancer, the National Surgical Adju- 
vant Breast and Bowel Project, NSABP, led by Dr. Bernard Fisher 
of the University of Pittsburgh. 

A particular focus of this hearing is the Federal Government's re- 
sponse to these problems. This case has major implications for 
thousands of patients who have participated in NSABP studies for 
decades, for the American taxpayers who have funded NSABP 
studies in amounts in excess of $100 million, and for maintaining 
the public trust in the integrity of science. 

Regrettably, this is only the latest in a series of cases that the 
subcommittee has felt it necessary to examine. The subcommittee's 
involvement in examining, investigating and monitoring the han- 
dling of cases of scientific misconduct dates back to 1988 when the 
subcommittee held its first hearing on the issue. 

Many in the scientific community have resisted outside scrutiny 
and others have sought to minimize the problem. But as we see 
today, scientific misconduct is a very real problem that requires an 
intensive and aggressive response by the scientific community itself 
and by the Federal Government. The case before us is a vivid re- 
minder of how poor the response of the scientific community can 
be and how serious consequences may be when the scientific com- 
munity and the Federal Government fall down on the job. 

It is, I think, without necessity that the Chair points out that in- 
volved in this matter is not only the confidence of the country in 
science, the questions of expenditure of public money, but the peace 
of mind of literally millions of women in the United States who feel 
great concern about the possibility of cancer and the consequences 
to them of bad science. 

Before we get into the specifics of this case, the Chair would like 
to point out that we will hear today important reassurances for 

(1) 



American women, and we believe that is good. According to the Na- 
tional Institutes of Health, the National Statistical Analyses by the 
National Cancer Institute, NCI, contractor indicate that the major 
findings of the NSABP lumpectomy/mastectomy study remain valid 
even without the inclusion of the fraudulent data. However, be- 
cause recent revelations have shaken confidence in the entire 
study, the Chair notes that audits are both necessary and under 
way to confirm the validity of the remainder of the data in the 
NSABP database. 

What this case shows, however, and what should continue to be 
of concern to all Americans are major problems in the handling of 
this matter by all concerned. Patient records were falsified and fab- 
ricated for over 13 years before NSABP's initial discovery in June 
of 1990 of data discrepancies at Montreal's St. Luc Hospital. There 
was no follow-up until September 1990, when Pittsburgh sent two 
staffers to Montreal, who not only confirmed this discrepancy but 
identified a number of others. In December one of the NSABP 
audit staffers wrote a memorandum describing that on two occa- 
sions St. Luc had continued to report follow-up on patients who 
had been dead for 6 to 8 months. 

I have had signatures on petitions which were filed against me 
by candidates of people who were dead, but I assumed that went 
on in politics and not in science. 

At the beginning of December 1990 Dr. Fisher told Dr. Poisson, 
who led the questioned work, that he had a data problem and that 
NSABP was sending a team to Montreal to investigate. Despite 
talking with the target of the investigation, Pittsburgh and Dr. 
Fisher still had not disclosed the problem to NCI. 

In February 1991 Pittsburgh sent its audit report to Dr. Poisson 
at St. Luc at which time he confessed to falsifying and fabricating 
records. About a week later Pittsburgh formally notified NCI of the 
fraud. Eight months had elapsed since NSABP's discovery of the 
fabricated and falsified data. The Office of Research Integrity, ORI, 
of the Department of Health and Human Services was notified and 
then began its investigation. 

In May 1991 Dr. Poisson admitted his fraud and was debarred 
for life by the U.S. Food and Drug Administration from performing 
research involving investigational drugs. 

On July 3, 1991, Dr. Samuel Broder, director of NCI, was briefed 
on the status of OKI's St. Luc investigation in which ORI had 
found "94 cases involved data falsification or fabrications which can 
be solidly documented." 

Dr. Broder concluded that the fraudulent St. Luc data should be 
removed, that all previously reported studies should be reanalyzed 
and the reports published on the reanalysis of the work. 
Reanalyses have dribbled out over time from Pittsburgh to NCI. 
However, to this date, some 3 years later, no reanalysis has been 
published. In fact, the New England Journal of Medicine has stated 
that it will not accept NSABP's reanalysis manuscript until the au- 
dits are completed. 

In addition. Dr. Broder concluded that no St. Luc's data should 
be included in any future publications. However, this was sum- 
marily ignored by Dr. Fisher. Dr. Fisher submitted at least 13 pa- 



pers that contained fraudulent St. Luc data, of which seven have 
now been published. 

Despite Dr. Broder's call for an immediate reanalysis, Pittsburgh 
did not present a reanalysis until March 1992. This reanalysis was 
a simple slide show for the NCI/ORI staff, claiming that the origi- 
nally published conclusions remained unchanged when the fraudu- 
lent data were removed. NCI/ORI accepted this account, and based 
on the assumption that the public health was not at risk for the 
next year, NCI/ORI insisted on a blackout of news about the inves- 
tigation. 

Between 1992 and 1994 NCI sporadically, and only half- 
heartedly, encouraged NSABP to complete a manuscript reporting 
the reanalysis, to significantly upgrade its audit procedure, and to 
establish a data monitoring committee. Pittsburgh moved forward 
at an equally halfhearted pace. 

After public disclosure of this debacle, Pittsburgh delivered its 
reanalysis to NCI in a matter of days. NCI's internal statistician 
had significant concerns about the adequacy. 

Recently NCI discovered NSABP was nearly 1 year behind in 
scheduling site visit audits and 3 years behind in submitting site 
visit reports to NCI. Among the audit reports that were submitted 
there were numerous instances in which significant discrepancies 
apparently were left without any investigation or other follow-up. 

For example, during an emergency site visit to Pittsburgh NCI 
uncovered a Pittsburgh audit report dated September 1993 which 
included evidence of serious irregularity at St. Mary's, another 
Montreal hospital. Despite the repercussions of the St. Luc's fraud, 
Pittsburgh did not report this new problem to NCI, apparently 
thereby contributing to the recent removal of Dr. Fisher as prin- 
cipal investigator. Last week ORI initiated a misconduct investiga- 
tion of this apparently falsified data. 

One of the reasons we are here today is that no one followed the 
direction of the director of the NCI. Top NCI officials ignored the 
director's instructions. Pittsburgh ignored the directions of its fund- 
ing institution. In fact, top NCI officials have complained to the 
subcommittee staff that they could not even get Dr. Fisher to re- 
turn their phone calls, let alone to take direction from NCI. 

This illustrates a central problem identified in numerous sub- 
committee investigations of scientific misconduct: Who is in charge, 
the NIH, which is the funding institution, or other funding institu- 
tions, or the prominent investigators? NIH's capability and willing- 
ness to manage and oversee federally funded research continues to 
be a key question, and millions of dollars are being spent on this 
kind of research. 

A separate but related matter to be discussed today underlines 
the importance of disclosing and aggressively responding to prob- 
lems. Dr. Fisher and Pittsburgh did not adequately investigate 
deaths involved in the use of tamoxifen in clinical trials under his 
supervision, nor did they notify NCI and ICI in a timely manner. 

Tamoxifen has been widely used in treatment studies for years, 
but only in the last 2 years has tamoxifen been given to healthy 
women in a prevention trial. At the time of the initiation of this 
study there were concerns about side effects, particularly uterine 



and other cancers, but not a single death at this time had been at- 
tributable to this risk. 

However, on June 25, 1991, the first tamoxifen-related uterine 
cancer death occurred. The death was reported to Pittsburgh on 
August 5, 1991, but Dr. Fisher claims it took him 2 years to deter- 
mine the cause of death. He told the staff of this subcommittee that 
he was unable to obtain the autopsy analysis from the hospital in 
his study. 

In fact, by October 1993 Dr. Fisher was aware of at least four 
uterine cancer deaths attributable to tamoxifen. Just last Friday 
the FDA and the manufacturer warned doctors about the increased 
risk associated with tamoxifen. The company has told the sub- 
committee staff that it first learned of uterine cancer deaths caused 
by tamoxifen when it was informed by NCI, not Dr. Fisher, in De- 
cember 1993. 

Both of these matters, the St. Luc false data and the delay in re- 
porting tamoxifen deaths, show that responsible audits and over- 
sight are critical to the maintenance of trust and to scientific 
progress. The challenge before us is to make the scientific process 
more open and more accountable without politicizing it or burden- 
ing it. 

In recent years the Congress has attempted to augment the gov- 
ernment's ability to combat scientific misconduct. For example, in 
1992 we passed the Generic Drug Enforcement Act, which debars 
from FDA application-related work those involved in falsifying 
drug testing data provided to the FDA. In 1993 we passed the NIH 
Revitalization Act, which requires the development of research in- 
tegrity, policy and regulations. 

These statutes help, but they do not substitute for administrative 
vigilance by NIH nor do they substitute for awareness and action 
by the scientific community, which has the first and, indeed, the 
principal responsibility of policing itself, a matter on which it must 
succeed or other measures will have to be taken by statute or by 
other action by the government. 

The witnesses today include a woman victimized by breast can- 
cer, representatives of concerned women's health groups, and offi- 
cials from HHS. These include the director of NIH, the director of 
NCI, the director of the Office of Research Integrity. The Chair an- 
nounces with regret that Dr. Fisher was unable to testify, citing 
health reasons. ICI/Zeneca, the manufacturer of tamoxifen, claimed 
inability to provide witnesses from the United Kingdom on a week's 
notice. The committee may be interested in hearing them on per- 
haps greater notice or perhaps a more interesting hearing at a 
later time. 

We thank the witnesses for appearing before us today and we 
look forward to their testimony. 

The Chair recognizes the distinguished gentleman from Colorado 
for an opening statement. 

Mr. SCHAEFER. I thank the Chair. I really appreciate the Chair 
holding this hearing this morning. 

Americans are perhaps the best informed people inhabiting the 
earth. CNN brings them news 24 hours a day, and the tele- 
communications industry's super highways transfer certain vital 
information at the blink of an eye. But when it comes to making 



medical decisions, Americans rely on the integrity of scientific re- 
search and the oversight functions of the Federal Government to 
guarantee that the information being received is accurate and com- 
plete. 

In this regard, the American people have been let down by the 
scientific community and representatives of the government. The 
facts of this case are well known. You can't pick up a magazine or 
a newspaper without seeing some mention of the fraud that was 
perpetrated by Dr. Poisson. 

While it is hard to believe that research scientists would delib- 
erately falsify data critical to an important study, it is even more 
illogical that his colleagues in the scientific community seem to 
have looked the other way. 

Perhaps worst of all is the fact that the National Cancer Insti- 
tute has disseminated incorrect and misleading information that 
has resulted in millions of American women wondering about the 
honesty and the accuracy of American science. 

Mr. Chairman, there is blame enough to be spread around here 
and it is important that we identify the individuals and institutions 
responsible for this disaster and take appropriate actions to assure 
that it does not happen again. More importantly, however, we must 
seek to reassure the American people about the soundness of Amer- 
ican science. 

I understand that, despite press statements to the contrary, the 
National Cancer Institute has still not conducted a comprehensive 
reanalysis of the study in question. This is unacceptable. Unaccept- 
able. NCI's tardiness and inaction does a disservice to the hard 
working members of the scientific community and to the American 
people who are sponsoring the research. 

Mr. Chairman, I want to commend you for holding this hearing 
and swiftly bringing this matter to everybody's attention. American 
women are scared, and rightfully so. That is why it is important 
that we do everything that we possibly can to allay their fears and 
restore their confidence in American research. I particularly am 
very pleased to see two colleagues of mine, Congresswoman Snowe 
and Congresswoman Schroeder, my colleague from the State of Col- 
orado, here today. I thank you, Mr. Chairman and yield back. 

Mr. DiNGELL. The Chair thanks the gentleman. 

The Chair recognizes now the distinguished gentleman from 
California, chairman of the health subcommittee, my friend, Mr. 
Waxman. 

Mr. Waxman. Thank you very much, Mr. Chairman. 

Public confidence in the research of the National Institutes of 
Health and particularly its National Cancer Institute are as impor- 
tant to scientific progress as the quality of its scientists or the level 
of its funding. Scientific misconduct, however rare, cannot be toler- 
ated and the impact, particularly upon clinical research, must be 
remedied quickly. 

Responsibility rests with the scientific managers of the NIH to 
be vigilant and alert to the potential for fraud and the implications 
for patient care. The response to knowledge of the falsified data 
from Canada was neither swift nor thorough. The consequence has 
been understandable concern on the part of breast cancer patients 
and their families. 



I am heartened that the recommendations of the National Cancer 
Institute regarding the use of breast-sparing procedures remain un- 
changed. I pray the NCI's continuing audit of the National Surgical 
Adjuvant Breast and Bowel Project database does not reveal addi- 
tional cases of misconduct and that the stronger administrative 
safeguards put in place will deter a repetition of this event. Our 
mothers, sisters and daughters deserve nothing less. 

Mr. Chairman, I want to highlight an additional concern that our 
subcommittee is reviewing in cooperation with Senator Rockefeller 
involving the tamoxifen prevention trial. Recent revelations regard- 
ing the risk of uterine cancer have resulted in stronger, more ex- 
plicit labeling changes for tamoxifen. 

I believe the NCI's decision to proceed with a prevention trial in- 
volving the administration of a potentially toxic drug to otherwise 
healthy women raises serious ethical questions. Hard questions 
need to be asked about the adequacy of the informed consent proc- 
ess. We need to carefully scrutinize the planning process used by 
the National Cancer Institute in initiating this trial. 

I look forward to working with you on this matter and I com- 
mend you for holding this hearing. 

Mr. DiNGELL. The Chair thanks the gentleman. 

The Chair recognizes now the gentlewoman from Illinois, Mrs. 
Collins. 

Mrs. Collins. Thank you, Mr. Chairman. As you know, I have 
been committed to fighting breast cancer by every means available 
to us for many, many years. As a matter of fact, I have pressed for 
better coverage for poor women by seeking to get preventive mam- 
mogram screening covered under Medicare for years, having au- 
thored the first such legislation way back in 1974. I supported ef- 
forts to educate the public about the prevention and early detection 
through breast cancer awareness months and other kinds of activi- 
ties that have gone on. 

I have also supported standards for quality mammograms and 
pressed for more attention and funding for research on women's 
health issues as a whole. 

As one who has been active on many fronts in the war against 
this disease, I must say that it is absolutely disheartening to wit- 
ness this episode. It is difficult enough to discover that researchers 
would falsify data that would be used in determining treatment 
regimens for breast cancer patients around the world, but to learn 
that every effort and precaution was not taken in rooting out the 
fraudulent data quickly and completely is truly tragic. 

Similarly, I am concerned that all speed was not used to inform 
the women who have breast cancer and who have the most to lose 
that there was a problem with this data. This is an even more trag- 
ic consequence. These actions have at the very least eroded public 
trust in the good efforts of the breast cancer research community 
to realize good treatments and preventions for this very deadly dis- 
ease. 

I am disappointed too that the women have been put through 
this emotional wringer and outraged that some of the scientific 
community have insensitively described their anguish as "fretting." 



I sincerely hope that this hearing will help us to uncover what 
went wrong at NSABP and at other points in the system and to 
determine what we can do to keep it from happening again. 

While it is clear that this episode has had an abominable effect 
on women confronting major decisions about treatment regimens 
for breast cancer, it is equally clear that researchers in the breast 
cancer research community who have contributed significantly in 
the past have sustained extensive damage to their professional rep- 
utations. I cannot help but wonder what effect the moratorium on 
enrolling new participants in clinical trials is going to have on the 
many other important studies that were being run by the NSABP 
or the chilling impact this controversy has on new scientists or doc- 
tors when they decide which research areas they want to study. 

Even so, I understand the importance of airing this issue, be- 
cause it is critically important, and I think it is the only way that 
we can regain confidence in the important work of the NSABP and 
the other government agencies that are involved. Without our over- 
sight, we could find ourselves in this situation again and, worse 
still, suspect the integrity of the scientific data on which women 
must make life and death decisions. 

Mr. Chairman, I am anxious to hear from today's witnesses and 
hope that through their testimony we can begin to rebuild some of 
the public trust in the way the government oversees medical re- 
search. I thank you, Mr. Chairman, and yield back the balance of 
my time. 

Mr. DiNGELL. The Chair thanks my good friend from Illinois for 
a very valuable opening statement. 

The Chair recognizes the gentleman from Oklahoma, Mr. Synar, 
for an opening statement. 

Mr. Synar. Mr. Chairman, thank you very much. I join with my 
colleagues today in discussing this very important subject. I think 
it is important that we try to relieve some of the tension and the 
emotion of this issue and, more importantly, to try to present a 
front that we are concerned about the research that is going on. 
Millions of women throughout this country are counting on this 
government and this Congress to make sure that this happens, and 
I join with you in this hearing. 

Mr. DiNGELL. The Chair thanks the gentleman for a very helpful 
opening statement. 

The Chair recognizes the gentlewoman from Pennsylvania, Ms. 
Margolies-Mezvinsky. 

Ms. Margolies-Mezvinsky. Thank vou, Mr. Chairman. Mr. 
Chairman, I want to thank you especially for holding this impor- 
tant hearing on behalf of all of the women and our families in this 
country, women faced with breast cancer, our daughters who may 
face this dreaded disease, and all of the families that must confront 
the painful decisions in treating this disease. 

In my district alone, Montgomery County, women face one of the 
highest rates of breast cancer in the State of Pennsylvania. Every 
family that has been sitting around at a dinner table and has had 
to make the difficult decision about what treatment to choose for 
our loved ones, be it mother, daughter or wife, deserves to know 
the truth and the findings behind these government financed stud- 
ies. When that trust is violated, like it was by the National Cancer 



8 

Institute and by the National Surgical Adjuvant Breast and Bowel 
Project, women and our families in this country have the right to 
be outraged. 

As important, we deserve answers as to why our own govern- 
ment misled us. Women now face even more fear and greater un- 
certainty because of these recent disclosures that data was falsified 
and fabricated. Our government must get its house in order and 
must properly ensure the absolute validity of its own studies. The 
women of this Nation deserve better, and I am eager to see that 
we get the facts about the dangers that confront us in order to 
make informed decisions about our lives in the future. 

I am particularly pleased to welcome Ms. Frances Visco from 
Philadelphia, who is president of the National Breast Cancer Coali- 
tion. 

Thank you, Mr. Chairman. 

Mr. DiNGELL. The Chair thanks the gentlewoman from Penn- 
sylvania. 

The Chair advises that pursuant to the rules of the committee 
it is within the discretion of the Chair to permit nonmembers of the 
committee to make opening statements, so the Chair is going to 
now recognize the two gentlewomen who are most interested in 
this in the House, Ms. Schroeder of Colorado and Ms. Snowe of 
Maine. The Chair will recognize first the gentlewoman from Colo- 
rado and then the gentlewoman from Maine. 

STATEMENT OF HON. PATRICIA SCHROEDER, A REPRESENTA- 
TIVE IN CONGRESS FROM THE STATE OF COLORADO 

Ms. Schroeder. Mr. Chairman, I thank you for letting us sit in. 
Congresswoman Snowe and I are here to thank you and thank my 
colleague from Colorado and everyone on this committee for calling 
this hearing. This committee is really one of the beacons of light 
that the women's caucuses look to. 

As I think back, it was about 4 years ago we were in this very 
hearing room when we first heard from the Government Account- 
ing Office that one more time NIH had not paid attention to its 
own guidelines and did not have any women in the research mod- 
els. We thank you for your constant, constant diligence in this area. 

But I must say, here we are, 4 years later, and you wonder if 
these researchers are ever going to catch on. This is a serious in- 
tegrity issue. It is about a disease that will kill 46,000 women in 
this country this year and afflict another 182,000, and yet some- 
body felt, ah, it wasn't that important; they could sit on this fal- 
sified data and not do anything with it, and, by the way, it didn't 
matter anjrway because they had some other data. 

Is it any wonder sometimes we women think that we are not get- 
ting a decent deal and that our health is not being treated with the 
same respect that other people's are? 

I was very pleased to hear Congressman Waxman's concerns 
about tamoxifen. One more time we find out about that, but there 
was a great delay from when they knew to when they did anything 
about it, and so I really have great questions whether this trial 
should continue. If you lured some women in who didn't sign the 
consent decrees and they knew it at that time and now they are 



going to keep the trial going, I think one more time it is saying we 
are playing fast and loose with women's health. 

I think it is time we say to these researchers there is only one 
reason I vote for money for medical research, and that is because 
I think good medical research is going to come around as good 
treatment; it is not to give them a little sandbox they can play in 
and do anything they want and not be accountable to anybody and 
never have to answer any questions. 

Somehow I thought this was the ego center of America, but I am 
beginning to think some of the scientific ivory towers are the ego 
center of America, because they get Federal money from those of 
us who give it to them and then they turn around and are abso- 
lutely offended that we would ever ask questions about the quality 
of it. If we can't get quality research that leads to quality treat- 
ment, then why are we even spending money for it?. 

There is absolutely no question that women are beginning not to 
trust anything the government does. For 4 long years this sub- 
committee, the women's caucus and many of us have been pound- 
ing away trying to get this changed, and for 4 long years it says 
a lot of people still haven't gotten the message. 

Thank you, Mr. Chairman. I'm sure we are all going to work to- 
gether to find some way we finally do give them the message or 
shut off their money, one or the other. I thank you for calling this 
hearing. 

Mr. DiNGELL. The time of the gentlewoman has expired. 

[The prepared statement of Ms. Schroeder follows:] 

Statement of Hon. Patricia Schroeder, a Representative in Congress from 

THE State of Colorado 

Mr. Chairman, thank you for allowing me to join you here on the dias today, even 
though I am not a member of this subcommittee. I'm breaking away from chairing 
a Research and Technology Subcommittee hearing, so I hope you'll excuse me if I 
give my opening statement and run. 

As co-cnair of the Congressional Caucus for Women's Issues and on behalf of 
women across the country, I want to thank you for taking women's health seriously. 
These hearings are a benchmark — thev signify the new standards of proof that 
women are demanding from research that will ultimately determine whether they 
live or die. 

Science has to be accountable to people. Mr. Chairman, because you have been 
so vigilant in being America's eyes and ears on scientific fraud, I applaud you. But 
you can't do it alone. We need people who are consumers of this information to have 
a role in how science is done. 

The concept of "checks and beilances" must extend to science. Women's experience 
with the scientific research community is proof of this. It was right here in this com- 
mittee room 4 years ago that the General Accounting Office released its report say- 
ing that the National Institutes of Health was not enforcing its own policy of includ- 
ing women in clinical trials. 

That revelation led to a revolution in women's health research. I believe that your 
work here today on the recent fraud in breast cancer research and the flaws in how 
scientific research is conducted in this country will lead to another revolution — one 
in which consumers are part of the decision-making process of how science is as- 
sessed. 

Consumers are not scientists. We know that. But we are not stupid, either. We 
don't want lies or sugar-coated results — -just information, good information. We do 
not want to be patronized, especially in issues of life and death. 

Let me turn to the specifics of this hearing. The National Surgical Adjuvant 
Breast and Bowel Project was looking into the survival rates of women with breast 
cancer who had lumpectomies or mastectomies. The project's 1985 study said 
mastectomies usually weren't necessary, which meant many women were saved 
from the disfigurement and trauma of having their entire breasts removed. 



10 

And then we found out that a Montreal doctor had falsified some data in the 
study, devastating the women who had undergone or were considering surgery, and 
casting a shadow over what little research there was on women's health. As if this 
werent enough, we learned that the head of the project, Dr. Bernard Fisher, was 
aware of this fraud since 1990 and that the National Cancer Institute knew about 
it since 1991. So, all this time passed without a published reevaluation of the find- 
ings. 

This is serious business, Mr. Chairman. We're talking about women's lives. We're 
talking about the integrity of the science behind treatment of a disease that will 
kill 46,000 women this year and afflict another 182,000. We're talking about re- 
search with a conscience. 

We're told the fraudulent data didn't affect the results of this and other studies. 
Even if that is true, it is unforgivable that women's health and well-being should 
have been treated so cavalierly, or that women should have been so needlessly 
frightened. 

None of this has been helped by the recent developments with tamoxifen, a drug 
used to treat breast cancer. Once again we encounter the National Surgical Adju- 
vant Breast and Bowel Project, Dr. Bernard Fisher, the NCI, and delays in inform- 
ing women about risks. And those risks, as we recently have learned, involve higher 
incidences of uterine and other cancers, and a possible link between the drug and 
a DES-like syndrome in fetuses. There have even been some deaths. 

The NCI and the National Surgical Adjuvant Breast and Bowel Project knew 
about this updated data last fall. But that data didn't get incorporated into the in- 
formed consent document for the prevention trial, which involves healthy women 
with no history of breast cancer, until mid-January of this year. There are still some 
women participating in this trial who have not been informed of these risks. It's 
questionable wnether the trial should even continue. 

The bottom line in all of this is that quality of research translates into quality 
of treatment — and the trust that people place in that treatment. And women have 
very little trust right now because their research has been compromised, resulting 
in too many confusing signals. 

Your role in uncovering scientific fraud and insisting science be accountable is 
critical. Giving consumers a role as well would help restore women's trust in the 
research they have funded with their tax dollars. And I think it would help save 
lives. 

Mr. DiNGELL. The Chair recognizes the gentlewoman from 
Maine, Ms. Snowe. 

STATEMENT OF HON. OLYMPIA J. SNOWE, A REPRESENTATIVE 
IN CONGRESS FROM THE STATE OF MAINE 

Ms. Snowe. Thank vou, Mr. Chairman. I want to join Congress- 
woman Schroeder in thanking you and Mr. Schaefer for inviting us 
to sit on the subcommittee today to discuss this very important 
issue to millions of American women, and I want to thank you for 
your timeliness in investigating this issue and the circumstances 
surrounding breast cancer fraud in the research. 

I have to say, as Congresswoman Schroeder said, we have been 
dealing with many lapses in women's health issues. We dealt with 
a lapse in funding, a lapse in enforcement policy that led to the 
systematic exclusion and discrimination against women in clinical 
study trials, and now we are dealing with an ethics lapse, a very 
serious lapse in a long string of lapses concerning women's health 
matters. 

I have to say that it just shatters our confidence in the medical 
and the scientific establishment and clearly undermines our trust 
in our ability to conduct accurate and fair clinical study trials, and 
it strikes at the very heart of women's fears as to whether or not 
the clinical study trial process which is the basis upon which we 
get information about breast cancer, has now been jeopardized. 

We spend billions of dollars. That is what we have been trying 
to do as the Congresswomen's Caucus, to increase the amount of 



11 

funding for breast cancer, and we have done that. We established 
an Office of Women's Health Research at NIH. We galvanized 
women all across America to be educated and aware of the need 
to have mammograms, and the medical community, and we have 
done that. Now we find out that we cannot entrust our health re- 
searchers with accurate information and reliable data. It truly is 
disheartening. 

I think women in America, at least in the response that we have 
had as the Congresswomen's Caucus to the efforts to upgrade wom- 
en's health funding and put it on a level playing field, finally saw 
the light at the end of the tunnel, that their needs were finally 
being addressed and not ignored. Well, that light at the end of the 
tunnel has sort of been snuffed out by this most recent disclosure. 

More disheartening than the fact that the falsified data was in- 
cluded in published studies is the way in which this issue has been 
handled by our primary research institute. As has been mentioned, 
it was discovered in June of 1990, but it took 8 months before a 
report was made by any project official. It took 8 months. And it 
has been 3 years since Dr. Poisson had admitted the fraud in the 
first place and 4 years since the discovery of the initial discrepancy. 
Four years. And it was just 1 month ago that the American public 
discovered this information. 

Why did it take so long to inform the public? Why has it taken 
so long for the National Cancer Institute to reanalyze this informa- 
tion when in fact they made a request for reanalysis in June 1992 
and January 1993? In fact, in October of 1992 NCI apparently ad- 
monished the project staff of their audit policy, noting that prob- 
lems with the St. Luc data went undetected for 13 years and that 
the project's audit policy contributed to the delay in detecting sig- 
nificant data irregularities. 

The real terrifying impact of this research fiasco ultimately is on 
the victims, those women who have breast cancer, because they are 
going to be haunted by the notion as to whether or not they have 
made a correct decision; they are going to question the data, the 
accuracy of the scientific inwrmation, their doctor's recommenda- 
tion. And who can blame them, Mr. Chairman? How can we allay 
their fears given this real atrophy of our institutions' medical re- 
search abilities, and who can really blame them for doubting what 
decisions they have made? 

In 1992, according to Newsweek magazine, 45,000 women chose 
to have lumpectomies and 117,000 chose to have mastectomies. It 
seems to me that these women deserve to know the truthful an- 
swers with respect to the choices that they have made. The women 
of America deserve no less. 

Thank you, Mr. Chairman. 

Mr. DiNGELL. The Chair thanks the gentlewoman. 

The Chair advises that we will now qualify our first panel. Our 
first panel is composed of Ms. Cynthia A. Pearson, program direc- 
tor. National Women's Health Network; Ms. Frances Visco, presi- 
dent. National Breast Cancer Coalition; and Ms. Jill Sigal. 

Ladies, if you will please come forward. The Chair would like to 
welcome each of you for your presence today and your assistance 
to us. It is much appreciated and we believe that your assistance 
will enable us to have not only a better record, but to move vigor- 



12 

ously forward to correct the concerns that I know you share with 
us today. 

As you know, it is the practice of the committee that all wit- 
nesses testify under oath. Do any of you have any objection to testi- 
fying under oath? 

[No response.] 

Mr. DiNGELL. The Chair advises in addition to this that it is your 
right to be advised by counsel if you testify under oath. Do any of 
you desire to be advised by counsel during your appearances here? 

[No response.] 

Mr. DiNGELL. The Chair advises also that copies of the rules in 
conformity with the Rules of the House, the Rules of the Sub- 
committee, the Rules of the Full Committee, the Rules of the House 
are there in the red and blue books available to you to inform you 
of your rights as you appear here and also of the limitations on the 
powers of the committee. 

Ladies, if you have no objection to testifying under oath, if you 
would please each rise and raise your right hand. 

[Witnesses sworn.] 

Mr. DiNGELL. You may each consider yourself under oath. The 
committee will recognize you in this order. First, Ms. Visco; second, 
Ms. Pearson; and third, Ms. Sigal. Again, the Chair thanks you for 
your assistance and cooperation with the committee. 

TESTIMONY OF FRAN VISCO, PRESIDENT, NATIONAL BREAST 
CANCER COALITION; CYNTHIA A- PEARSON, PROGRAM DI- 
RECTOR, NATIONAL WOMEN'S HEALTH NETWORK; AND JILL 
LEA SIGAL, CONSULTANT, WASHINGTON, DC. 

Ms. VisCO. Thank you, Mr. Chairman and members of the En- 
ergy and Commerce Subcommittee on Oversight and Investiga- 
tions. I am Fran Visco. I'm a breast cancer survivor, I'm a wife and 
mother, and the president of the National Breast Cancer Coalition. 
I was diagnosed with breast cancer in 1987 and I chose to have a 
lumpectomy rather than a mastectomy based on my own independ- 
ent readings and the information and advice given to me by physi- 
cians. 

On behalf of the National Breast Cancer Coalition and the 2.6 
million women who are living with breast cancer in this country 
today, I want to thank you for your work on breast cancer and for 
this hearing today. It is an extremely important happening. 

A diagnosis of breast cancer carries with it a never-ending sense 
of unease, of concern and fear. Because of my own diagnosis and 
my work with the National Breast Cancer Coalition, my life is peo- 
pled with women who have heard the words, 'Tou have breast can- 
cer", with their families and friends. We are everyone. We are law- 
yers, we are scientists, we are Members of Congress, we are teach- 
ers, homemakers, mothers and daughters. We are, by and large, a 
medically sophisticated vocal group who live day in and day out 
with the threat, the fear and the pain of breast cancer. 

About 3 years ago many of us took that fear and we turned out- 
ward into a national advocacy movement, and we formed the Na- 
tional Breast Cancer Coalition. Our mission is to eradicate the 
breast cancer epidemic. Our goals are to increase research and 
funding, to ensure access for all women, and to increase the influ- 



13 

ence of women with breast cancer and other consumer advocates in 
the decision-making that impacts our lives. 

Recently the anxiety and fear with which we live daily has been 
exacerbated by a barrage of events. First, the unfortunate manner 
in which the National Cancer Institute announced its decision to 
change the guidelines for mammography; then the news belatedly 
announced fraud in the lumpectomy studies; and finally, the uter- 
ine cancer deaths from tamoxifen. 

No one at the National Breast Cancer Coalition has been un- 
touched by these recent events at NCI and NSABP. Mr. Chairman, 
we must do more to ensure that these systems that allow this 
fraud to occur, that allow the information to be kept from the pub- 
lic, are held responsible and corrected. 

While we appreciate the fact that it appears as though this time 
women's health in general and women with breast cancer in par- 
ticular were not placed in great jeopardy by the deceit and neg- 
ligence of the scientists involved, we are outraged that the informa- 
tion was kept from the public, from women with breast cancer, 
from women who had decisions to make, and from other scientists 
and other scientific institutions, from the physicians who were ad- 
vising women. 

With a diagnosis of breast cancer we find ourselves in a world 
over which we have very little control. We have to learn a new sci- 
entific language. And we do. We have to understand new concepts. 
And we do. And we are asked to turn ourselves over unthinking, 
unquestioning to the scientific and medical communities. We do 
not. Not any longer. We insist on being part of the decision-making 
that impacts our lives. 

The recent barrage of news about breast cancer research and 
treatment underscores the urgency and necessity of our demand 
that consumer advocates, that breast cancer advocates have a seat 
at the table. 

The discovery that officials at NCI and NSABP not only knew 
about the fraud but that they did nothing to make this important 
information readily available to the public and the scientific com- 
munities is the most disturbing fact. This failure to respond has 
caused advocates to question the level of trust that the public 
places in these institutions. After the dust has cleared from all the 
media attention, the congressional inquiries there will still remain 
a crisis of confidence among millions of women who have been 
asked to trust the institutions charged with acting in the best in- 
terest of our health. 

I recognize that Dr. Bernard Fisher was a visionary in the field 
of breast cancer and that his leadership in the area was in part re- 
sponsible for the advances that have been made in this disease, 
and I also recognize the tensions and difficulties faced by the Na- 
tional Institutes of Health and the National Cancer Institute. 

I have received a number of telephone calls over the past weeks 
from members of the scientific community. They want to make cer- 
tain that we activists understand that we must not, in their words, 
"Throw the baby out with the bath water", that we must not insist 
on controls that are so severe that the scientific process is unneces- 
sarily impeded. We know. We understand. But we still must ask 
these questions. What procedures were followed? Did they conform 



14 

with accepted standards? And if they did, are the accepted stand- 
ards sufficient? And most critically, if the status quo is such that 
scientific fraud is kept from the public, from other scientists and 
from our physicians, then the status quo must change. 

We understand the need for large clinical trials. We support 
them; we advocate for them; but we need change. In this situation 
nothing changed. This information has been available for quite 
sometime but nothing changed until activists and advocates spoke 
out, until the Congressional Caucus for Women's Issues spoke out, 
until you, Mr. Chairman, and this committee spoke out and held 
these hearings. 

We know we must be clear for American women. There are other 
studies that support the results of the NSABP studies. There are 
benefits to tamoxifen. And if we have all of the information we 
need, we can make informed decisions. 

I have been told not to overreact, not to question too much. We 
didn't perform fraud. We didn't neglect to uncover fraud for 13 
years. We didn't delay reanalysis of data. We didn't keep informa- 
tion from the public. The scientific community did that, and we 
have every right and indeed a responsibility to call them to task 
for this. I'm sitting here. I'm a member of the President's Cancer 
Panel; I'm a breast cancer survivor; I lead a national breast cancer 
organization; and I found out about this by picking up the New 
York Times. And I still don't know what happened, what was 
reanalyzed, when it was, who did it, who is involved. 

The public trust in the system is eroding and we are going to get 
it back through hearings like this and by letting consumers be part 
of the decision-making process. 

This sorry story reveals a system overly concerned with profes- 
sional reputation and institutional ego, both public and private in- 
stitutional ego. And while the participants shuffle to position them- 
selves to best protect themselves and to point a finger at someone 
else, I ask them to stop and to look at me, look at all the women 
in this room, all the women in your lives. It is my life; it is our 
lives that your decisions impact. It is my money; it is public money 
that you spend. We women with breast cancer, consumer advocates 
belong at the table. We must be a part of the decision-making, of 
data monitoring committees, of oversight committees, of study sec- 
tions. 

There is resistance in the scientific community to this. We want 
to "micromanage their research." We "don't understand." "Science 
is pure." But we are here spending much time and money to fix im- 
pure science and bad decisions. Science is not a concept that exists 
in a vacuum. It is performed too often in isolation, insulated from 
the public by individuals, by imperfect people with biases and 
shortcomings. That is why we need a strong system of checks and 
balances, oversight, different perspectives that will allow science to 
proceed unimpeded by ego, fraud and erosion of public trust, and 
that will occur if consumer advocates are given an informed, mean- 
ingful seat at the table. 

I recently read in the paper a quote from a top official of NCI 
who said if a year or two ago he had been able to intuit women's 
concerns about the NCI and NSABP behavior, NCI would have 
acted differently. Well, there is a fundamental deficiency in a sys- 



15 

tern where a public servant believes he must intuit how the public 
must react to decisions made by his agency. Had a consumer advo- 
cate, a woman with breast cancer been part of the process the 
public's peace of mind and women's lives would not have been left 
to the uncertainties of an individual's intuitive ability. This self-im- 
posed separation between public agencies and the public they serve 
is unacceptable. 

We are gathered here today because of a crisis, a fundamentsd 
flaw in the systems that are supposed to protect us. I feel very 
strongly, and I know you share that, that we would do a larger dis- 
service to the women of this country if we did not use this as an 
opportunity to ensure this does not happen again. Let's make it a 
matter of policy that consumers, breast cancer advocates are in- 
volved at every stage of the research process, from advisory boards 
to study sections at the agency level, to steering committees, data 
monitoring committees and IRB's at the institutional and study 
levels. We have to be a part of this. 

I want to thank you, Mr. Chairman and members of this sub- 
committee. Your work has brought us closer to reestablishing pub- 
lic confidence in our institutions and the scientific process, but I 
know that I do not want this to be the method of oversight of 
science. 

Thank you. 

Mr. DiNGELL. Ms. Visco, thank you for your very helpful testi- 
mony. We appreciate your presence and your assistance. 

Ms. Pearson. 

[The prepared statement and attachment of Ms. Visco follow:] 



16 



1^1 National 
r^^ Breast C: 



_ Cancer 

± Z Coalition 

a grassroots advocacy effort 



PO Box 663"'} 
Washington. DC 20035 

(202) 296-7477 Testimony of Fran Visco 

President, National Breast Cancer Coalition 

before the 

Subcommittee on Oversight and Investigations 

April 13, 1994 



Thank you, Uli. Chairman and members of the Energy and Commerce 
Subcommittee on Oversight and Investigations. I am Fran Visco, a breast cancer 
survivor, a wife and mother, a lawyer and President of the National Breast Cancer 
Coalition. 

On behalf of the National Breast Cancer Coalition and the 2.6 million women 
who are now living with breast cancer, thank you for your work on the pressing issues of 
breast cancer and for the opportunity to testify before this Subcommittee. 

A diagnosis of breast cancer carries with it a never-ending sense of unease, of 
fear, of concern. Because of my own diagnosis and my work with the National Breast 
Cancer Coalition, my life is now peopled with women who have heard the words, "you 
have breast cancer", with their families and friends, with women and men who have lost 
loved ones to this disease. We are everyone - we are lawyers, scientists, Members of 
Congress, teachers, bomemakers, mothers and daughters. We are, by and large, a 
sophisticated, vocal group who live day in and day out with the threat, the fear and the 
pain of breast cancer. 

About three years ago, many of us took that fear and turned it outward, into a 
national advocacy movement - the National Breast Cancer Coalitioa The Coalition is a 
grassroots advocacy organization dedicated to the eradication of the breast cancer 
epidemic. We are made up of more than 270 organizations and thousands of individual 
women, their families, friends and health care providers. 

Our mission is to eradicate the breast cancer epidemic. Our goals are to increase 
funding for breast cancer research and to help focus the research on prevention and 
cure, as well as treatment and screenings; to ensure access for all women to health care 
and to increase the influence of women living with breast cancer and other breast cancer 
advocates in the decision making that impacts their lives. 



17 



Recently the anxiety and fear, with which we live daily has been escalated by a 
barrage of events: first, the unfortunate manner in which the National Cancer Institute 
announced its decision to change the guidelines for mammography screening; then, the 
news, (belatedly announced), of fraud in the lumpectomy studies; and finally, the uterine 
cancer deaths from the tamoxifen trials. 

No one at NBCC has been untouched by the recent events at NCI and the 
National Surgical Adjuvant Breast and Bowel Project. Mr. Chairman, we must do more 
to ensure that the systems that allowed this fraud to occur are held responsible and 
corrected. We are very disturbed to learn the extent of the fraud. While we appreciate 
the fact that it appears as though this time women's health in general, and women with 
breast cancer in particular, were not placed in great jeopardy by the deceit and 
negligence of the scientists involved, we are outraged that the information was kept from 
the public, from other institutions, and from health care providers. 

With the diagnosis of breast cancer, we find ourselves in a world over which we 
have little control. We must learn a new scientific language - we do; we must 
understand new concepts - we do; we are asked to turn ourselves over, unthinking and 
unquestioning, to the scientific and medical communities - we do not, not any longer. I 
know that many members of the scientific community - and most of those in power - do 
not believe that we belong at the table. They tell us their science is "pure" and that our 
involvement will somehow interfere with their work. They accuse us of wanting to 
"micromanage" research. The scientific community does not understand: it is not our 
intent to micromanage what they do or to interfere with the process of research. We 
have simply learned that when the process is left solely to the scientists, women are not 
well served. The recent barrage of bad news about breast cancer research and treatment 
underscores the urgency and necessity of our demand that breast cancer advocates, 
including women with breast cancer, have a seat at the table. 

We are here today because, time after time, it is unclear whether women's health 
is truly at the center of concern. Certainly Doctor Roger Poisson's fraud is indefensible. 
What really concerns me though, is how we deal with these problems when they arise. 

The discovery that officials at the National Cancer Institute and the National 
Surgical Adjuvant Breast and Bowel Project not only knew about the fraud, but that they 
did nothing to make this important information readily available to the public is the most 
disturbing faa. This failure to respond to the breach of scientifically valid research 
techniques has caused breast cancer advocates to question the level of trust which the 
public places in these institutions. After the dust has cleared from all of the media 
attention and congressional inquiries, there will remain a crisis of confidence among 
millions of women who have been asked to trust the institutions charged with acting in 
the best interest of our health. 

The lumpectomy study at NSABP is an enormous undertaking -- it involves 
hundreds of international sites. Granted, as the size of any study increases, it becomes 
more and more difficult to exercise absolutely tight controls. But the problem in this 



18 



case is that when it became clear - years ago -- that the data from this landmark study 
were false, nothing was done to ensure that women's health was protected. Opportunity 
after opportunity to correct the false data were lost. It is hard to believe that there was 
ever any real intent to ensure that women and their doctors were aware of the problems 
in the study. It frightens me to think when, or if, we would have ever learned the truth if 
the Chicago Tribune hadn't done its own research. 

Unfortunately, the lumpectomy studies aren't the only thing imder scrutiny today. 
The Food and Drug Administration released information last Friday about the dangers 
of the drug tamoxifen which has been used successfully in women after surgery for breast 
cancer. Unfortunately, tamoxifen has also been shown to increase the risk of uterine 
cancer. Six women who were taking tamoxifen to prevent the recurrence of disease are 
dead of uterine cancer. Currently, healthy women are enrolled in tamoxifen trials to 
determine if it is an effective preventive agent for breast cancer. Thousands of women 
are still enrolled in this trial at unknown risk to their lives. I am concerned that once 
again, efforts will not be made to inform the public about the risks associated with this 
drug. When NCI develops and tests a drug successfully, there is a large publicity effort, 
but when studies show unexpected risks, somehow that information gets bogged down in 
procedure and protocols. Consumers - women with breast cancer - are the last to 
know. 

Mr. Chairman, we must ensure that women are informed in a timely manner 
about the risks they assume when participating in any kind of scientific study. 
Presumably, lumpectomy studies and tamoxifen trials are undertaken in the best interests 
of women and their health. But as research goes forward, the best interests of women 
become secondary to scientific protocol, to bureaucracy, and to preservation of 
professional reputations. 

I have received several messages over the past week from members of the 
scientific community. The scientists want to make certain that I understand that we 
should not, in their words, "throw the baby out with the bath water"; that we must not 
insist on controls that are so severe that the scientific process is unnecessarily impeded. 
We know. We know. But we still must ask these questions: What procedures were 
followed? Did they conform with the accepted standards? And if they did, are the 
accepted standards sufficient? Most critically, if the status quo is such that scientific 
fraud is kept from the public, from other scientists, from our physicians, then the status 
quo must change. 

I recognize that Dr. Bernard Fisher was a visionary in the field of breast cancer. 
And that his leadership in this area was in part responsible for the advances that have 
been made. I also recognize the tensions and difficulties faced by the National Institutes 
of Health and the National Cancer Institute. But when we concentrate too much power 
in too few people, we do a disservice to the public. I have seen what individual, 
professional, and institutional ego can do: those in power get to a point where they 
become insulated from the public and from the patients they serve. 

We are gathered here today because of a crisis - a fundamental flaw in the 
systems that are supposed to protect us. But I feel very strongly that we would do a 
larger disservice to women if we do not use this as an opportunity to ensure justice in the 
future. We must make it a matter of policy that consumers - breast cancer advocates - 
are involved at every stage of the research process, from advisory boards to study 
sections at the agency level, to steering committees; data monitoring committees and 
IRB's, at the institutional and study level. Women must be a part of the research 
process if their best interests are to be fully ensured. 

Thank you Mr. Chairman and members of the Subcommittee. 



19 



(i^B National 
r-^^ Breast Cancer 
• ^* Coalition 

^T Al^l a grassroots advocacy effort 

NATIONAL BREAST CANCER COALITION STATEMENT 
P o Box 66373 jj^ RESPONSE TO NSABP DATA FALSIFICATION 

Wuhingioa. DC 2003 S 

(202) 296-7477 

A tenet of the National Breast Cancer Coalition policy agenda is that consumers - breast 
cancer advocates - belong at every level of the research process, from advisory boards to 
study sections at the agency level, to steering committees, data monitoring committees and IRBs, at 
the instinitional and study level. To advance this agenda," for the past three years, the Coalition has 
met with top officials at the National Institutes of Health and the National Cancer Institute, and 
demanded meaningful representation at every level. 

Recent events at the National Cancer Institute and the National Surgical Adjuvant Breast and 
Bowel Project (NSABP) underscore the necessity of including consumer advocates at the table. NCI 
and NSABP knew for several years of falsification of data that were used in critical breast cancer 
smdies of the efficacy of lumpectomy for certain breast cancers and the efficacy of tamoxifen in 
preventing recurrence and death from breast cancer, and failed to reveal this information to the 
public. In fact, a manuscript submitted by NSABP to the NCI journal in January, 1994 included 
results from an investigator known to have fabricated data. 

In response to the public's outcry that information has been kept from it, a top official of 
NCI recently was quoted as saying that if a year or two ago he had "intuited" women's concerns to 
the NCI/NSABP behavior, NCI would have acted differently. There is a fundamental deficiency in a 
system where a public servant believes he must intuit how the public may react to decisions made by 
his agency. Had a consumer advocate - a woman with breast cancer - been pan of the process, the 
public's peace of mind and women's lives would not have been left to the uncertainties of an 
individual's intuitive abilities. This self-imposed separation between the public agencies and the 
citizenry they serve is unacceptable. ^^^ ' 

The behavior of the NCI and the NSABP have put women's lives at risk. These recent 
revelations have undermined the effectiveness of the NCI and NSABP and have caused breast cancer 
advocates to question the level of trust to which these institutions are entitled from the public. 

The public funds biomedical research; scientists who perform this research are accountable to 
the public. NCI and NSABP's failure to disclose these data in a timely manner impinges upon the 
rights of the patient who is entitled to make her decisions with all relevant information revealed, of 
the scientist and physician who have a responsibility to the process and their patients, and of the 
public that funds the research. 

The vast amounts of money spent on biomedical research warrant structures that do not 
consolidate power in the few, but rather foster diversification. Diversification will result in new 
ideas and approaches and decreased opportunities for development of a "club-like" atmosphere 
among researchers. This will help to build a system of checks and balances and minimize the 
possibility of vital information being kept from the public. 

The National Breast Cancer Coalition demands an independent investigation of both the 
underlying falsification of the data submitted to the NSABP and of the failure of NCI and the 
NSABP to timely disclose information to the public. We demand that the names of the 
biostatisticians named as investigators by the NCI be revealed immediately and that the nature of 
their charge be made public at once. 



20 

TESTIMONY OF CYNTHIA A- PEARSON 

Ms. Pearson. Good morning, Chairman Dingell, members of the 
committee. Thank you for inviting us to participate in this impor- 
tant hearing. I represent the National Women's Health Network. 
The Network is a strong supporter of clinical trial research for 
women. We remember all too vividly the harm that women suffered 
when they were treated on the basis of hope or belief rather than 
scientific knowledge. 

When we have a context of that kind of commitment to clinical 
trial research, we were shocked to find out that those involved in 
conducting and overseeing breast cancer treatment trials have 
withheld information about falsified data. We were angry when we 
learned from the press that the National Cancer Institute and 
NSABP delayed for over 1 year after a final report from ORI docu- 
menting falsified data in NSABP. We were angry when we learned 
that no reanalysis had been done. In the past NCI has rushed 
NSABP's results to the public when there was good news. To delay 
announcing falsification of data within breast cancer trials makes 
it appear tnat Dr. Fisher and NCI want the public to know results 
only if they reflect well on NSABP and NCI. 

This disregard for the public's right to know is outrageous. We 
commend Congress for stepping in when it appears that NCI isn't 
able to assert its authority. Dr. Fisher was seemingly unable or un- 
willing to follow NSABP's own guidelines for auditing of data and 
timely reporting of possible problems. NCI apparently took none of 
the appropriate measures that would have ensured compliance. 

Dr. Fisher's failure to report problems with data in the breast 
cancer treatment trials and NCI's failure to require compliance ap- 

Eear to be paralleled by their actions with regard to the tamoxifen 
reast cancer prevention trial. In this prevention trial important 
information that volunteers and the general public need to know is 
withheld for unreasonably long periods of time, and when eventu- 
ally given to women, it is incomplete and misleading. 

By withholding information about the effects of the drug used in 
the prevention trial, NCI and Dr. Fisher are attempting to sidestep 
a public discussion of the wisdom of continuing this trial. 

The breast cancer prevention trial, originally a bad idea, is well 
on its way to becoming a disaster. The trial was a bad idea to begin 
with because it wasn't designed to prevent disease — breast can- 
cer — in healthy women. Because it is using a drug — ^tamoxifen — 
that is known to have serious side effects, it's an attempt to sub- 
stitute one disease for another. And now, because of the recent rev- 
elations, we find that the risks of tamoxifen are actually much 
worse. 

We believe that if these results which have been withheld were 
openly disclosed when the trial began that it would not have been 
approved. We also believe that the new information is enough jus- 
tification to stop the trial. To let this trial proceed is to watch 
women die of trust. 

We would like to briefly review for you the chronology of misin- 
formation by omission and commission in which Dr. Fisher and 
NCI have collaborated. 

In April 1992, when the trial began, healthy volunteers were 
given a consent form that was misleading in two very important 



21 

ways. It told them the risk of an increased chance of uterine cancer 
was about three. That risk was based on old data, which Dr. Fisher 
and NCI well knew more cases had been reported. It also told 
women, "No deaths from uterine cancer were reported." 

In fact, as you said in your introduction, as early as 1991 Dr. 
Fisher and NSABP knew that a tamoxifen treated breast cancer 
patient who developed uterine cancer had died, but instead of 
warning women of this possible risk, the consent form told women 
that uterine cancers that had occurred were thought to be curable. 

After the trial began. Congress attempted to exercise some over- 
sight. Congressman Ted Weiss chaired the Human Resources and 
Intergovernmental Relations Subcommittee. They held an inquiry 
and a hearing later in 1992. NIH, led by that time by Dr. Healy, 
resisted all efforts for that committee to exercise its oversight into 
the quality of consent that was taking place at the trial and into 
the scientific basis for conducting the trial. The trial continued as 
planned. 

Throughout 1992 and 1993 reports of different kinds of problems, 
new kinds of problems with tamoxifen emerged and were reported 
in the scientific literature. These reports described liver damage, 
eye problems, skyrocketing estrogen levels in pre-menopausal 
women, cancers of the gastrointestinal tract, and aggressive 
endometrial cancers. 

Through 1992 and 1993 NCI and Dr. Fisher were strangely si- 
lent about the results of the important large tamoxifen breast can- 
cer treatment trial called B-14. That trial had reported in 1991 
originally no uterine cancers associated with treating breast cancer 
patients with tamoxifen. Well, what was happening behind the 
scenes in 1992 and 1993 was frightening. These researchers knew 
that from zero the number of uterine cancers had jumped to 15 in 
the tamoxifen arm of the trial, and 4 of those women had died of 
their uterine cancers. Throughout all this time women were not no- 
tified, doctors taking care of them were not notified, and the 
healthy women volunteering to participate in the prevention trial 
were not notified. 

Finally, in November 1993 Dr. Fisher informed NCI of the 
deaths. Over 2 more months passed before Fisher and NCI in- 
formed women in the prevention trial. We have calculated that 
over 600 women were given a consent form that said there were 
no deaths from uterine cancer after the NCI had been informed 
about deaths. 

This is very disturbing, but unfortunately it isn't even the worst 
of the misleading information that has been given to women by Dr. 
Fisher and NCI. Women were given an information update in Jan- 
uary 1992. It told them that the increased risk of breast cancer was 
approximately threefold compared to similar women in a general 
population. 

Well, in fact, on April 6 Dr. Fisher published his study upon 
which those numbers were based and told the scientific community 
in the NCI's own journal that the risk he found was 7.5 times as 
high. Women in the prevention trial have never been told that. 
They still aren't being told that to this day. They are being given 
a guesstimate, and a guesstimate that is far lower than the results 



22 

of a randomized, placebo controlled, double blind trial, the scientific 
gold standard for evidence. 

I ask the committee to consider, if any of you were volunteering 
for a research trial, how would you feel if you were told that a risk 
of a complication was increased by three times and you later found 
out that a good scientific study had found a risk of 7.5 times? 

This pattern of behavior has resulted in the general public being 
subjected to a smoke and mirror show that prevents a reasoned 
discussion of whether the trial should proceed. 

Because of this behavior on the part of Dr. Fisher and NCI, we 
have come to the state where Congressional oversight is badly 
needed. In addition to the types of changes that Ms. Visco has de- 
scribed, which we agree with completely, we also call on Congress 
to ensure that there is an appropriate review of the scientific basis 
for the tamoxifen prevention trial, and any appropriate review will 
have to be conducted by an agency other than the NCI and be com- 
posed of a majority of prevention experts, public health experts, not 
cancer treatment doctors. 

The National Women's Health Network believes that this trial 
should be stopped, the tamoxifen prevention trial should be 
stopped, and that any objective, independent review will come to 
the same conclusion. 

Peter Latham, a 19th Century physician said that, "Medicines 
and poisons are oftentimes the same substance given with different 
intents." 

To women with breast cancer tamoxifen is good medicine. To 
healthy women it may be closer to poison. 

Thank you. 

[The prepared statement of Ms. Pearson follows:] 



23 



Cynthia A. Pearson 

Program Director 

National Women's Health Network 

Adriane Fugh-Berman, MD 

Medical Advisor 

National Women's Health Network 

Good morning. Chairman Dingell and members of the Committee. Thank you for inviting 
us to participate in this important hearing. 

The National Women's Health Network is a non-profit women's health advocacy group. 
We are financially supported by our membership of 400 local women's health projects and 
over 17,000 individuals. Our mission is tvvo-fold: to advocate for better federal health 
policies for women, and to provide women with accurate, useful information which gives 
them more power in decisions about their health care. 

As a natural outgrowth of those goals, the Network has been a strong supporter of clinical 
trial research for women. We remember all too vividly the harm that women suffered when 
they were treated on the basis of hope or belief, rather than scientific knowledge. The use 
of DES in healthy pregnant women during the 1950s and 1960s and the routine use of fetal 
electronic monitoring in low risk women in the 1980s took place because clinical trials 
either were not done, or were ignored. Women and their babies were hurt, not helped, by 
these medical interventions. 

In addition to our concern that all aspects of women's medical care be based on well- 
founded scientific research, the Network has been particularly concerned about women's 
breast cancer treatment. In the 1970s, surgical treatment for breast cancer was a high 
priority issue for the newly emerging women's health movement. Radical mastectomy, and 
the "one-step" biopsy/mastectomy procedure were questioned by activists and authors such 
as Rose Kushner and Barbara Seaman. American women learned that lumpectomy was an 
option for women diagnosed with breast cancer who lived in other countries. When U.S. 
physicians resisted women's requests for lumpectomy, women turned to the research 
establishment and demanded studies that would convince surgeons that lumpectomy was 
safe. 

Those studies began in the U.S. and other countries in the late 1970s. The researchers 
brave enough to withstand the disdain of the general surgical community were much 
applauded by women's health activists. The largest of these trials was coordinated by Dr. 
Bernard Fisher, head of the National Surgical Adjuvant Breast and Bowel Project (NSABP). 



Dr. Fisher was hailed as a hero by many in the women's health movement as well as the 
cancer survivor community. Now, nearly twenty years later. Dr. Fisher is the same man 
who has been found to demonstrate such utter disregard for the rights of patients to know 
the results of research. It may seem ironic that we have come today to criticize the 
behavior of a researcher whom previously we praised. However, it is vitally necessary that 
the government act decisively to establish clear standards to protect the public's right to 
swift and full disclosure of all important research findings, whether good, bad, or 
embarrassing. 

As a consumer group our biggest complaint about the behavior of Dr. Fisher and the 
National Cancer Institute (NCI) is the delay of over a year in aimouncing to the public that 
falsified data were submitted to NSABP. Dr. Fisher and NCI also delayed releasing a 



24 



reanalysis of previously published studies. In the past, NCI has rushed Dr. Fisher's results 
to the public, even before publication, when there was good news. To delay announcing 
the falsification of data within the breast cancer trials makes it appear that Dr. Fisher and 
NCI want the public to know results only if they reflect well on NSABP and NCI. 

This disregard for the public's right to know is outrageous. We commend Congress for 
stepping in to oversee when it appears that NCI isn't able to assert its authority. Dr. Fisher 
seemingly was unable or unwilling to follow NSABP's own gwdelines for auditing of data 
and timely reporting of possible problems with data. NCI apparently took none of the 
appropriate measures that would ensure compliance. 

Keeping consumers in the dark has been a theme of the tamoxifen prevention trial fiasco. 
Dr. Fisher's failure to report problems with data in the breast cancer treatment trials in a 
timely way, and NCI's failure to require compliance with standard guidelines for auditing 
and reporting appear to be paralleled by their actions with regard to the breast cancer 
prevention trial. We see the same pattern of behavior. Important information that 
volunteers in the trial and the general public need to know is withheld for unreasonably 
long periods of time, and when eventually given to women it is incomplete and misleading. 
This pattern of behavior has important consequences for women's health, and important 
public consequences as well. By withholding information about the effects of the drug used 
in the prevention experiment, NCI and Dr. Fisher are attempting to sidestep a public 
discussion of the wisdom of continuing this trial. 

The breast cancer prevention trial, originally a bad idea, is well on its way to becoming a 
disaster. It is designed to give healthy women tamoxifen, a breast cancer treatment drug, at 
the same dose taken by patients with cancer. The trial accepts any woman over the age of 
35 whose risk of developing breast cancer in the next five years is 1.7 percent. At the time 
this trial began, the known serious risks of the drug were reported to be almost exactly 
equal to the volunteers' risk of developing breast cancer. Specifically, uterine cancer and 
blood clots occurred in 1.8 percent of women who took tamoxifen for five years. The trial 
was a bad idea to begin with because it wasn't designed to test whether tamoxifen could 
prevent disease in healthy women, but rather to determine whether one disease could be 
substituted for another. 

Now, we find that the risks of tamoxifen, known to be serious at the time the trial began, 
are actually much worse. We believe that if the results which have been withheld by Dr. 
Fisher had been openly disclosed when the trial began that it would not have been 
approved. We also believe that the new information is enough justification to stop the trial. 
To let this trial proceed is to watch women die of trust 

We would like to briefly review the chronology of misinformation by omission and 
commission in which Dr. Fisher and NCI have collaborated. 

In April, 1992, when the tamoxifen prevention trial began, women were given a consent 
form that told them the risk of uterine cancer was increased by about three times based on 
existing data from several large trials of tamoxifen. Women \vere told that nine out of 



25 



3,097 women on tamoxifen developed uterine cancer versus four out of 3,091 women not 
treated with tamoxifen. These numbers were based on a 1991 report prepared by an NCI 
scientist. Dr. Susan Nayfield, who relied upon data supplied by the manufacturer to the 
FDA in 1990. By 1992, NSABP had reports of four more tamoxifen-treated women who 
had developed uterine cancer. NSABP included this information in the prevention trial 
protocol, but did not include these new cancers in the consent form that women were given. 
If the most up-to-date information had been given to women, they would have been told 
that the risk of uterine cancer was not three times as likely, but four or five times as likely. 

In addition, the 1992 consent form assured women that "no deaths from uterine cancer were 
reported". Again, the consent form was seriously misleading. As early as 1991, Dr. Fisher 
and NSABP apparently knew that a tamoxifen-treated patient who developed uterine cancer 
had died. But instead of warning women of this possible risk, the consent form told 
women that the uterine cancers that had occurred were "thought to be curable". 

The 1992 consent form was misleading in other ways, as well. Women were told that a 
very small number of women taking tamoxifen might die as a result of blood clots. What 
they weren't told was that NSABP had found that a much larger number of women suffered 
life-threatening blood clots which required hospitalization, and indefinite treatment with 
blood-thinning medication. Also, women were told that liver cancer had been reported in 
rats receiving tamoxifen in doses greater than the dose used in humans. Women were not 
told that the manufacturer of tamoxifen had reported to the FDA in a public hearing that 
the cancer-causing dose of tamoxifen in rats produced levels of tamoxifen in the blood that 
were equivalent to the blood levels of women taking tamoxifen in all of the NSABP trials. 

After the trial began, we were deeply concerned that women were being hoodwinked into 
volunteering for a trial that put their health at risk. We turned to Congress for help, and 
worked with Congressman Ted Weiss, who chaired the Human Resources and 
Intergovernmental Relations Subcommittee of the Committee on Government Relations. 
The Committee conducted an inquiry into the quality of the informed consent forms used in 
the tamoxifen prevention trial and found shocking results. Two hundred and sixty eight 
forms were examined. Sixty eight percent either omitted or altered one or more of the key 
points from the model consent form approved by NCI reviewers. To put it bluntly, NCI 
and NSABP had allowed a bad consent form to become worse. Although Mr. Weiss died 
as the inquiry was completed, his conunittee held hearings chaired by Congressman Donald 
Payne in October, 1992. At the hearing, the Committee members attempt^ to exercise 
their oversight function by questioning the underlying scientific basis of the risks, the 
possible benefits of tamoxifen, and the poor quality of information given to women 
volunteering for the trial. NIH was represented by Dr. Healy at that hearing. Dr. Healy 
resisted all efforts of the committee and the eventual outcome of the hearing was that a few 
consent forms which didn't even meet minimum legal requirements for informed consent 
were improved. But the trial continued as originally planned. 

Since the last time Congress reviewed the tamoxifen prevention trial, more evidence about 
its harm has been published. Liver damage in women on tamoxifen has been reported to 
regulatory bodies in both the U.S. and Britain. Depression, eye damage and increased rates 



26 

of gastrointestinal cancers (including liver cancer) have been reported. Problems with 
ovarian function in premenopausal women, including skyrocketing estrogen levels and large 
ovarian cysts, have also been reported. 

As these reports accumulated, NSABP and NCI were silent as to any update from B-14, the 
large tamoxifen treatment trial coordinated by Dr. Fisher. The original results had been 
published in 1989. No updated information was provided throughout 1992 and 1993, either 
to the general public, medical journals, or those who received the prevention protocol. We 
have now discovered that what was happening behind the scenes was frightening. Uterine 
cancers were being diagnosed at a rapid rate, and more women were dying as a result. 
From zero uterine cancers at the time of the 1989 publication, to six as of May, 1991, the 
number jumped to 1 5 in the tamoxifen treated arm of the study. And still women weren't 
notified. 

Finally, in November, 1993, Dr. Fisher informed NCI of the increased incidence of uterine 
cancer in breast cancer patients taking tamoxifen, and of the four deaths from uterine 
cancer. Even after this long delay. Dr. Fisher and NCI allowed months to pass before 
informing women. Throughout all of 1992 and 1993, women in the prevention trial were 
being given a consent form that falsely reassured them that uterine cancer was not fatal. 
Over two months passed after Dr. Fisher's report to NCI before women were informed 
through an information update. We have calculated that nearly 600 women were given a 
consent form that said there were no deaths from uterine cancer after NCI had been 
informed about the deaths. 

This is very disturbing, but unfortunately, it isn't the worst of the misleading information 
that has been given to women by Dr. Fisher and NCI. The information update given to 
women in the prevention trial in January, 1994 accurately informed women that women 
taking tamoxifen had died from uterine cancer. However, it also told women that the risk 
of developing uterine cancer was three times higher than a similar group of women in the 
general population. 

It wasn't until April 6, when Dr. Fisher's uterine cancer paper was fmally published in the 
Journal of the Nationd Cancer Institute that we found out that the risk of developing 
uterine cancer for women in the NSABP trial was not three times as high, but actually 
seven and one half times as high. In his discussion. Dr. Fisher attempted to minimize his 
own results. With NCI's consent, women in the prevention trial were told a "guesstimate" 
of their risk for uterine cancer that was far lower than the results of B-14. A "guesstimate" 
that is nowhere close to the risk found in a randomized, placebo controlled trial ~ the gold 
standard of scientific evidence. We would ask the Committee members to consider whether 
if they were volunteering for a research trial, they would want to be told that the risk of a 
serious complication was increased by three times, when a good study had found an 
increased risk of seven. 

We can't impute motive to Dr. Fisher and NCI, but it appears as if they don't like the 
results of the NSABP trials, and so they are trying to avoid others becoming aware of 
unpopular results. Similar to NCI's acquiescence. Dr. Fisher's unwillingness to publish a 



27 



reanalysis of the NSABP trial affected by falsified data, NCI has allowed Dr. Fisher to 
present women with extremely misleading information about the results of the B-14 trial. 

This pattern of behavior has resulted in the general public being subjected to a smoke and 
mirrors show that prevents a reasoned discussion of whether this trial should proceed. NCI 
and Dr. Fisher have also refused to respond to scientists with legitimate critiques of the 
risks and possible benefits of the trial. Last fall, two important reviews of the prevention 
trial were published by credible scientists in peer-reviewed journals. Neither critique has 
received a response. To consumer advocacy groups it appears as if Dr. Fisher and NCI 
don't believe that along with public trust, and public tax dollars, comes a duty to participate 
in public discussion. 

In summary, the National Women's Health Network is convinced that Congressional 
oversight is necessary in this case. We hope that there will be two results from this effort 
We call for an increased commitment by NCI to timely disclosure of all important 
information from clinical trials. We also call for the establishment of a system to ensure 
monitoring and follow through of that commitment. 

We also call on Congress to ensure that there is an appropriate review of the scientific basis 
for the tamoxifen prevention trial. Any appropriate review will have to be conducted by an 
agency other than NCI and include a majority of prevention and public health experts, not 
cancer treatment researchers. The National Women's Health Network belives that this trial 
should be stopped and an objective, independent review will come to the same conclusion. 
Peter Merc Latham, a 19th century physician, said "medicines and poisons are oftentimes 
the same substance given with different intents". To women with breast cancer, tamoxifen 
is good medicine. To healthy women, it is closer to poiscm. 



28 

Mr. DiNGELL. Ms. Sigal, we are happy to welcome you here 
today. The time I last saw you at the fund-raiser that we were par- 
ticipating in I was not aware we were going to have you before the 
committee. I notice you look a little bit uncomfortable. I want you 
to know that the committee is known for having sharp teeth, but 
we are also known for being careful on whom we use them. So I 
hope you feel comfortable and welcome. 

TESTIMONY OF JILL LEA SIGAL 

Ms. Sigal. Thank you, Mr. Chairman. I've been in the room 
when you have exhibited your sharp teeth, so I'm glad I'm not 
going to get that today. 

Mr. Chairman, Mr. Schaefer, members of the subcommittee, my 
name is Jill Sigal. I reside in Alexandria, Virginia, and I work in 
Washington, DC., as a consultant. I am 32 years old and just 6 
months ago I was diagnosed with breast cancer. My doctors in- 
formed me at that time that I had a choice in surgical procedures. 
I could either have a lumpectomy with follow-up radiation or a 
mastectomy. 

Loss of a breast is a difficult prospect for any woman. Being 32, 
loss of a breast was an especially frightening option. However, if 
the choice was between having one breast or accepting a substan- 
tially higher risk of premature death, the choice for me would have 
been a clear one. I would have opted for life. 

I spent the 2 weeks between the time of my biopsy and the time 
of my surgery talking with numerous doctors at various hospitals 
all over the Washington area trying to gather information in order 
to make an informed decision. My doctors did not recommend a 
particular procedure to me. They left the decision up to me. My 
doctors laid out the facts for me and they told me about one study 
that concluded that a lumpectomy with follow-up radiation had the 
same long-term survival rate as a mastectomy. The study that my 
doctors cited to me is the study that is in question today that we 
now know includes falsified data, as admitted by Dr. Poisson. 

I had a lumpectomy and I based my decision to do that solely on 
the study that is in question today. Since my surgery I have under- 
gone 6V2 weeks of radiation and I am currently in the middle of 
chemotherapy. I am halfway done. 

I learned only 2 weeks ago that Dr. Poisson had falsified data 
and therefore the results of the study are being questioned. I did 
not learn about this fraud from the National Cancer Institute; I did 
not learn about this fraud from the National Surgical Adjuvant 
Breast and Bowel Project; I did not learn about this fraud from the 
University of Pittsburgh or from Dr. Fisher. I learned about this 
fraud because one day I was looking over a friend's shoulder who 
happened to be reading the New York Times and a headline 
jumped out at me talking about the breast cancer fraud. 

I understand now that there was a notice in the June 21, 1993, 
Federal Register regarding Dr. Poisson. I went back this week and 
I read the notice. The notice doesn't even identify the study in 
question. Therefore, if I had read the Federal Register notice back 
in 1993, which I am not in the habit of reading the Federal Reg- 
ister, I would not have known what it was referring to because it 



29 

didn't mention the lumpectomy study. Reading the notice back in 
1993 would have proved to be of no use to me. 

My anger and outrage that a doctor could possibly engage in 
such gross scientific fraud was surpassed only by my disbelief when 
I learned that the National Cancer Institute, an instrument of the 
United States Government, knew about the falsified data 3 years 
ago and deliberately did not give it wide-spread publicity. 

I also understand that the University of Pittsburgh knew about 
the falsified data 3 years ago and deliberately did not give it wide- 
spread publicity. I also understand that the University of Pitts- 
burgh knew about the fraud approximately 4 years ago, and it 
withheld the information from the National Cancer Institute for a 
period of time. These organizations did not inform the public. They 
did not even inform the doctors who could have then advised their 
patients accordingly. 

How many women during these 3 years made a decision about 
their surgery, as I did, based on this study. How many women 
must now wonder, as I do every day, if they will die because they 
may have made the wrong decision? How many women, Mr. Chair- 
man, will die? We will never know. 

If my cancer returns, it is likely to come back in my liver, my 
lungs, or my bones. My doctors tell me that if this happens, my 
cancer cannot be cured and I will die. If this happens, my family 
and friends will never know if my cancer had spread prior to my 
operation and there really wasn't much that anybody could do or 
whether the reoccurrence was due to the fact that I had a 
lumpectomy instead of mastectomy. 

I take no comfort, no comfort whatsoever from the fact that the 
Institute that swept the fraud under the rug for 3 years now claims 
to have conducted a re-analysis of the study and maintains that 
the findings are still valid. For me, the National Cancer Institute 
has forfeited any claim to credibility. Indeed, I understand that 
when the National Cancer Institute first announced to the public 
that it had conducted a re-analysis, the announcement was false. 
It was only later, from my understanding, that a re-analysis was 
actually conducted, and even then the so-called re-analysis did not 
include a review of the raw data. 

Mr. Chairman, there seems to be no end to the fraud and decep- 
tion. As a result of the National Cancer Institute's behavior in this 
manner, I now question other policies of the National Cancer Insti- 
tute, including its policy that women under the age of 50 should 
not get a mammogram unless they are at high risk. 

Well, Mr. Chairman, I don't think there's a doctor in the world 
that would say that I was at high risk. In fact, the doctors that are 
treating me have told me that my chance of getting breast cancer 
at age 32 was 0.2 percent. My cancer had been growing inside me 
for several years. If I had had a mammogram a year or two ago, 
I would have possibly caught the cancer at its earliest stage when 
it's contained within the duct, and the survival rate would have 
been for me then approximately 98 percent. Well, my cancer was 
not caught at this stage. I had invasive cancer, and my chance of 
survival is far less than 98 percent. If I had had a mammogram 
earlier, I very likely would not be facing the daily fear of pre- 
mature death. 



30 

Mr. Chairman, there is nothing that this subcommittee can do to 
help me. My fate is cast, but there are several things that this sub- 
committee can do under your leadership to prevent similar atroc- 
ities from occurring in the future. I'm no expert in this area of 
breast cancer, but I ask the subcommittee to consider the following 
recommendations, some of which, based on your opening statement, 
Mr. Chairman, I understand may be currently underway. 

First, I ask the subcommittee to consider commissioning an inde- 
pendent re-analysis of the study and the raw data with a written 
report to be submitted to the subcommittee within 60 days. The re- 
port should be published and made available to the public. 

Two, order the release of the raw data with the names of the pa- 
tients redacted. 

Three, prohibit the National Cancer Institute from awarding any 
grants to the University of Pittsburgh for 5 years, or until after a 
thorough study of its procedures and safeguards has been con- 
ducted. 

Four, extend the 8 year debarment of Dr. Poisson from receiving 
Federal grants to a lifetime prohibition. 

Five, change the way allegations of scientific fraud are inves- 
tigated by the National Cancer Institute and the Federal Office of 
Research Integrity, including requiring timely widespread notifica- 
tion of scientific fraud. 

The fear that arises from facing one's own mortality at my age 
can at times be paralyzing. Now, as a result of this fraudulent 
study, and the apparent cover-up by the National Cancer Institute 
and the University of Pittsburgh, my terror is exacerbated. Today 
there exists a crisis of confidence and credibility as it pertains to 
the National Cancer Institute and to the other Organizations associ- 
ated with this study. Even if the conclusion of the study, Mr. 
Chairman, holds up after a proper re-analysis, think about the 
agony of uncertainty that I and thousands of others are currently 
enduring. 

I thought I had made an informed decision, and I thought I had 
done everything in my power to increase my chances of enjoying a 
normal life expectancy. Now, I must wonder every day if I really 
have done everything to maximize my chances of survival. 

If the members of this subcommittee can save just one life, then 
in my very humble opinion, you will have accomplished a lifetime 
achievement as a member of the House of Representatives. That 
life you save may be somebody you know. I do not understand how 
I got breast cancer, but I've come to accept it, and I deal with it 
as best I can. What I cannot understand and cannot accept is why 
a medical doctor would falsify data in a study that was intended 
to guide thousands in making what could be a life or death deci- 
sion, and why the people given the responsibility to oversee the 
study did not publicize the fraud immediately. 

I would like and need answers to these questions, Mr. Chairman. 
To me, this hearing is not just about breast cancer. To me, this 
hearing is about accountability. The people have a right to have 
faith and confidence in U.S. government sponsored research. These 
recent events show us clearly that the process has failed. Mr. 
Chairman, you and your committee have the authority and the 



31 

power to diminish or eliminate the possibility of this atrocity from 
ever happening again by changing and strengthening the process. 

Mr. Chairman, Mr. Schaefer, thank you from the bottom of my 
heart for holding this hearing and for doing what you can to correct 
this grave injustice. 

Mr. DiNGELL. Thank you, Ms. Sigal, for your most helpful state- 
ment. Be assured that we intend to pursue this matter with the 
usual vigor that this subcommittee has displayed over the years, 
and I would say that there are a number of people out there who 
probably will rest poorly being aware of that information. I will ob- 
serve parenthetically that we will anticipate hearing from the Uni- 
versity of Pittsburgh. I know they're looking forward for a chance 
to come forward and tell their side of the story, as we are looking 
forward to a chance to hear their side of the story and perhaps ask 
them a few simple little questions. 

Ms. SiGAL. Well, I take great confidence in that, Mr. Chairman. 
Thank you very much. 

Mr. DiNGELL. Well, be assured we will follow this matter forward 
vigorously. 

Ms. Sigal. Thank you, sir. 

Mr. DiNGELL. Your statement has been most impressive, and we 
thank you for your assistance. 

Ms. Sigal. Thank you, Mr. Chairman. 

Mr. DiNGELL. The Chair is now going to recognize members of 
the committee in the order prescribed by the rules. The Chair rec- 
ognizes first the gentleman from Colorado, Mr. Schaefer. 

Mr. Schaefer. Thank you very much, Mr. Chairman. I do appre- 
ciate the testimony of all three members before the committee 
today. We know that a dreadful situation has developed, not only 
with the people involved but the Cancer Institute, and I think 
we're all very interested in getting to them a little bit later on. I 
would ask Ms. Sigal, you testified that you first learned of the fal- 
sifications of the research data from news reports. Is that right? 
Your doctor did not know? 

Ms. Sigal. No. I was reading over a friend's shoulder who was 
reading the March 27 edition of the New York Times, and this 
headline, "Federal Officials to Review Documents in Breast Cancer 
Study, jumped out at me. That's the way I found out about it. I 
went to my doctor after I read this and I asked him if he knew. 
He found out about it the same way, reading about it in the paper. 

Mr. Schaefer. Ms. Visco? 

Ms. Visco. Mr. Schaefer, that's exactly how I found out about it, 
and I think 

Mr. Schaefer. By the newspaper? 

Ms. Visco. By the newspaper. I think it's particularly egregious, 
you know, my organization in October presented the President with 
2.6 million signatures on a petition from women and men and chil- 
dren across this country asking for a national action plan for breast 
cancer, and the President said yes. At his request, Secretary 
Shalala hosted a conference of the National Institutes of Health on 
December 14 where we had consumer activists, government rep- 
resentatives, scientific community representatives, private indus- 
try, the media, and representatives from Congress, come together 



32 

to sit at a table and begin the design of a national action plan, and 
we shared ideas and facts and concepts. 

I feel quite foolish today thinking back on that day because I 
thought we were there at the table on a level playing field, and I 
find out there was very important information that was kept from 
us. If it was ever going to be told to us, even belatedly, it should 
have been that day, and it was not. So yes, I found out about it 
from the newspaper. 

Ms. Pearson. Mr. Schaefer? 

Mr. Schaefer. Yes? 

Ms. Pearson. I read the Federal Register each day. Maybe I'm 
one of the few, but it's part of my job. I look at the table of contents 
and then I read the notices specific to women's health, and even 
I couldn't find it. That notice in June was buried. There was no 
way any reader could tell its importance. 

Mr. Schaefer. Don't you think — is there any responsibility on 
the physicians themselves to look at this? I mean, your doctors, 
they should have been at least told of what was going on, it seems 
to me. How can they advise you, even though they give the final 
decision to you on whether you have a lumpectomy or a mastec- 
tomy? Shouldn't this data be submitted to them on a continual 
basis? 

Ms. SiGAL. Absolutely, Mr. Schaefer. I mean, without this infor- 
mation being given to the doctors, the doctors can't advise people 
like myself adequately. I mean, my doctors did a great job. I'm very 
proud of the doctors that are caring for me, and they gave me the 
best advice they could and they laid out the facts as we knew them 
back then in October, just 6 months ago. If they had had this infor- 
mation, I would have seriously considered whether or not the risk 
was worth it of going through with a lumpectomy, and I probably 
would have gone forward. As difficult as it would have been emo- 
tionally for me, I probably would have had the mastectomy in order 
to maximize my chances of living, because that's what this is all 
about. 

Mr. Schaefer. Ms. Visco, how long was it ago that you have a 
lumpectomy? 

Ms. VisCO. September 1987. 

Mr. Schaefer. And you have gone back and been checked over 
this period of time? 

Ms. Visco. Yes. 

Mr. Schaefer. And there hasn't been any recurrence or any- 
thing? 

Ms. ViSCO. I have not been diagnosed with a recurrence, that's 
correct. 

Mr. Schaefer. Well, we're all glad of that. I know you are. 

Ms. Visco. Yes. 

Mr. Schaefer. But again, it's a decision that you made at that 
time, in 1987, and so it's worked out all right. 

Ms. VisCO. I understand the studies. I understand that there are 
independent studies that support the results of the NSABP spon- 
sored trial. I understand that. I think most women with breast can- 
cer do understand that, but we do live every day with the threat 
of a recurrence hanging over our heads because we don't know how 
to cure this disease. That level of concern is with me every single 



33 

day, and now on top of that, I have a new level of concern. It's un- 
necessary. It doesn't need to be there. This information — I'm sure 
there's fraud. I mean, you cannot guarantee that there's never 
going to be fraud, but once you find out about it, do something 
about it. 

Mr. SCHAEFER. I think this is right. My time is about up here, 
Mr. Chairman, but I think Ms. Sigal hit it right on the head. We're 
not only talking about this particular study and what the Cancer 
Institute has withheld or who has withheld what. We're talking 
about trust in the government on what information they dissemi- 
nate to the people. I think this is the bottom line that chairman 
wants to get at. I too would like to do that. So, I certainly appre- 
ciate your testimony here today, and we'll do all that we can in 
working with the chairman to make sure this does not ever happen 
again. 

Mr. DiNGELL. Will the gentleman yield to the Chair just very 
briefly? I just want to maJce a comment. We have gone through 
these questions of scientific misconduct, fraud and things of that 
kind. One of the things that's been most comforting to the Chair 
has been the wonderful, cooperative way with which all of my col- 
leagues on this subcommittee have cooperated with the Chair in 
these investigations, particularly the gentleman from Colorado. It 
has been — we have had great difficulties, as the gentleman very 
well recalls. The committee has been under attack for having in- 
sisted that the question of scientific misconduct and scientific fraud 
be addressed internally inside the scientific community, and also 
that Federal agencies address their question and the responsibility. 

We've also dealt with it up until now from the standpoint of the 
concern that we had and the jurisdictional responsibility we had 
about the funding with Federal dollars. Today, because of the com- 
ments of Ms. Sigal, Ms. Visco, and Ms. Pearson, we're seeing it now 
from a more personal standpoint, from the hurt, the danger, the 
peril, the trauma that it occasions to citizens who, interestingly 
enough, are taxpayers who support these programs, who are enti- 
tled to openness and truth, who are entitled to honorable and prop- 
er behavior by investigators, and proper discipline within the sci- 
entific community and within the government itself. Regrettably, 
we're not seeing that, but my comments to the gentleman particu- 
larly are my gratitude to him and the members of the committee. 
My commendations to him for the faithful and vigorous and decent 
way in which he has stood with the Chair as we've gone through 
some very hard times in these investigations. So, I express my 
thanks to the gentleman at this time, as I do to the rest of the 
members of the committee. 

The Chair recognizes now the gentlewoman from Illinois for 
questions. 

Ms. Collins. Thank you, Mr. Chairman. Ms. Sigal, I must say 
that your testimony was most impressive, and one that I don't 
think I'll ever forget. I certainly can't say that of a great many oth- 
ers that I have heard, not on this subject but on any subject as a 
matter of fact. Let me say that I think that you have done the 
country, and particularly those of us who are interested in this 
matter, a tremendous favor by coming to us and giving us your per- 
sonal testimony on what has happened as a direct result of the in- 



34 

formation that you did not have to avail yourself of regarding a 
treatment that you could have chosen, and I personally thank you 
for that. We all have a debt of gratitude that we owe you. 

The chairman has already said that we're going to have people 
in from Pittsburgh to answer some questions about why this was 
not revealed to you and to many other people, and we're also going 
to receive testimony at this hearing and the others from the NIH 
and the NCI. They would have us to believe that the issue here is 
simply probably that the conclusions of the lumpectomy/mastec- 
tomy trial are still valid. Do any of you at the table think that the 
conclusions are still valid? 

Ms. Pearson. This is interesting, Ms. Collins, because in the 
70's, lumpectomy was something that women demanded. We heard 
that it was available to women in other countries. They seemed to 
be surviving as long, but American surgeons resisted. So we, and 
some in the cancer research community called for trials, and that's 
why I opened by saying what a strong supporter we were of re- 
search. We were delighted that the lumpectomy trial began in the 
United States, and when it reported the results, we joined the Na- 
tional Cancer Institute in encouraging women to consider 
lumpectomy. We know that there were five other trials conducted 
independent of NCI and NSABP, some in other countries, that 
have found similar results. 

So, if the NCI and NSABP had had the sense and just the de- 
cency to come forthrightly immediately and say we found this 
fraud. It's horrible. We'll make sure it never happens again, here's 
our re-analysis. We think our trial is still the same, and there's 
these five other trials. Women would have had some worry, of 
course, but it could have been handled much more usefully if we 
had gotten the whole package of information all at once. Now it 
comes out as if there's a cover-up, and it's hard for women to even 
get the information that there are several other trials that show 
lumpectomy as effective as mastectomy. 

Ms. Visco. I echo what Ms. Pearson said. I also want to add that 
I think the fact that fraud was going on for 13 years and the fact 
that whatever procedures were in place were unable to detect that, 
and the fact that the NCI wasn't aware of that for quite some time, 
that has to be put into the mix also. Although we do recognize that 
there are other studies reporting the result, the concern in the pub- 
lic trust is affected. When you look at that and say well, what else 
is out there? If these are the procedures and they didn't work here 
and they got caught here, what else is out there that we don't know 
about? We need to do something about these procedures. 

Ms. Collins. Exactly. Ms. Sigal? 

Ms. Sigal. Mrs. Collins, first of all, thank you very much for 
those kind remarks. I think it's premature to say whether the con- 
clusion of the lumpectomy study is still valid. I personally have a 
big problem with anything the National Cancer Institute has said 
and probably will say today. It's not going to comfort me. As Ms. 
Visco said, this has been going on for 13 years. I think what needs 
to be done is we need to have a proper re-analysis of the study. I 
don't take their word for it. I can't because my life is depending on 
this, as well as hundreds of thousands of other women. So, for 
them to tell me, oh, you're going to be OK, that doesn't mean any- 



35 

thing to me today. I don't think it means anything to my sister 
who's sitting behind me or my mother, who's watching. No, we 
need to have a proper re-analysis and a review done. 

As regards to tne other studies, when I made my decision, I 
never knew that other studies existed because my doctors, I as- 
sume, felt that the study in question today was the pre-eminent 
study and that is the only one that they told me about. It wasn't 
until last Friday when I went racing to my doctor to talk to him 
about all of this, when I went to have my chemotherapy treatment, 
that he said well, Jill, there are other studies that exist. Well, but 
I share Ms. Visco's view. Well, if it could happen in one study, 
maybe it could happen again. I don't want to believe that. I do 
want to have faith in what government sponsored research tells us, 
but I think there's a credibility gap. I think with your help and the 
chairman's and this committee's, that you could fix that. 

Ms. Collins. Thank you. Ms. Visco, NCI and HHS officials have 
expressed their concerns about due process and fairness. They be- 
lieve it was best to not tell the public about the unreliable St. Luc's 
data until after the conclusion of the ORI investigation. Have you 
thought about whether HHS could have preserved fairness while 
also honoring the public's right to know and what strategies might 
HHS have pursued in this regard? 

Ms. VisCO. Yes. As a breast cancer survivor and as an attorney, 
I did give that quite a bit of thought. The process should not have 
taken that long. The information could have been released to the 
public much earlier. Simply, the investigation should not have 
taken that long. Dr. Poisson admitted falsifications. In addition, 
the NCI had the information and the investigation was complete, 
I believe the latest date I've heard is last spring, the spring of '83. 
We're just finding out about it today, and from the newspaper. So, 
I do believe that whatever process is in place to investigate has to 
be shortened, and I'm certain that it can be. 

Ms. Collins. Thank you. Thank you, Mr. Chairman. 

Mr. DiNGELL. The Chair thanks the gentlewoman. The Chair rec- 
ognizes now the gentlewoman from Pennsylvania, Ms. Margolies- 
Mezvinsky. 

Ms. Margolies-Mezvinsky. Thank you, Mr. Chairman. Ms. 
Sigal, today we hear testimony that in 1992, NCI and HHS officials 
failed to recognized that Dr. Fisher had not re-analyzed his data 
showing the recurrence of a tumor in the ipsilateral breast for la- 
dies that had had a lumpectomy. But these same NCI/HHS officials 
have told subcommittee staff that the absence of re-analvsis of this 
data was not important because occurrence of ipsilateral tumors is 
a secondary variable, a minor consideration they said. As a women 
who just 6 months ago had a lumpectomy, what's your reaction to 
this kind of thinking? 

Ms. Sigal. Well, it may not be important to the people in the Na- 
tional Cancer Institute, but I can tell you it is certainly important 
to me and to my family and my friends, who don't want to see me 
die because of this. So, my reaction to your statement is, I think 
that's an outrageous thing for the National — and inhumane thing 
for the National Cancer Institute to say. It's that kind of attitude 
that I think has led to this problem. I think people's egos have got- 
ten in the way of scientific integrity in research, and I believe that 



36 

there has to be some kind of reviewability for these scientists. If 
there's nothing to hide, then why not open up the process? 

Ms. Margolies-Mezvinsky. What did your doctor say to you? 

Ms. SiGAL. My doctor tried — I was there on Friday when I had 
my treatment, and I was quite upset. He said that there are other 
studies. Like I was telling Mrs. Collins, I did not know about that 
at the time when I made my decision to have a lumpectomy in Oc- 
tober, but he said that there are other studies and that everything 
is going to be OK. Well, I have great faith in my doctor, I really 
do, but I would greatly appreciate if we could have a proper re- 
analysis because until that is done, until it is done by some entity 
that's not connected with the National Cancer Institute, yes, I will 
still worry. My family will still worry, and I'm sure that thousands 
and tens of thousands of women who are out there in the same po- 
sition that I am will worry as well. 

Ms. Margolies-Mezvinsky. Did your doctor say anything about 
Dr. Fisher? 

Ms. SiGAL. Yes, he did. He was actually very complimentary of 
Dr. Fisher. I personally can't speak to Dr. Fisher. I don't really 
know much about what he has done except for what I've read in 
the papers, and it does sound very good, but yes, my doctor was 
very complimentary of him. In fact, he even said, "We don't need 
to throw the baby out with the bath water." He did say that. I 
looked at him and, quite frankly, while I was getting shot up in the 
veins during my chemotherapy treatment, we had quite a heated 
discussion about this whole subject for the hour that I was there. 
I'm not sure in this particular instance that I agree completely 
with what my doctor thinks in this instance, but I do have faith 
in the way he's treating me. 

Ms. Margolies-Mezvinsky. If Dr. Fisher and his colleagues 
were here today, is there any message that you would like to share 
with them over and above what you said in your testimony? 

Ms. SiGAL. I'm not sure there's a message that I'd like to send, 
but there are certainly questions that I would want to ask Dr. 
Fisher, like why. I want to know why. I want to know why the 
process broke down. I want to know why the procedures weren't 
followed, their own procedures, the procedures of the National Can- 
cer Institute and the National Institute of Health. I want to know 
why the audits weren't conducted over the last year because I don't 
understand. Is there some scientific reason? Did somebody's ego get 
in the way? I don't know, but that's what I'd like to know from Dr. 
Fisher and the University of Pittsburgh. Why did he allow this to 
happen? 

Ms. Margolies-Mezvinsky. The other panel members? 

Ms. Visco. I would like to speak for a moment to an issue that 
you raised earlier, and that is the fact that there apparently were 
some findings of recurrence but not mortality, and they were not 
considered statistically significant. This also speaks to the position 
that my organization takes, and I know that Ms. Pearson's organi- 
zation also supports, and that is we women with breast cancer and 
consumer advocates should be involved because I have been to a 
number of meetings where the scientific community has spoken 
about what they're looking at is the mortality rate. 



37 

We keep raising the issue of well, what about recurrence? What 
about quality of life? What about these issues? Mortality, how 
many women live and died is not the only question we should be 
looking at. In this trial, let's look at the women who had 
lumpectomies rather than mastectomies. How many recurrences 
were there? Those are not questions that the scientific community 
have asked. If women were at the table, those questions would be 
asked. 

Ms. Pearson. I guess what I would say to Dr. Fisher is that he 
seems to have lost the understanding that with public funds and 
public trust needs to come a commitment to public discussion. As 
I described in my testimony, there's been an apparent pattern of 
behavior of believing that Dr. Fisher, NSABP and to some extent 
the NCI, can decide for themselves what the public needs to know 
and what they're willing to talk about with the public. I would just 
want to let Dr. Fisher know that we disagree strongly with that. 

Ms. Margolies-Mezvinsky. Thank you. Mr. Chairman, I yield 
back the balance of my time. 

Mr. DiNGELL. The time of the gentlewoman has expired. The 
Chair recognizes now the gentleman from Ohio, Mr. Brown. 

Mr. Brown. Thank you, Mr. Chairman. Ms. Pearson, is it pos- 
sible for NCI/HHS to regain the trust of the public, particularly the 
trust of women at risk for and diagnosed with breast cancer? 

Ms. Pearson. I think it is, yes, because if they can't we're in big 
trouble. We'll never have scientific grounding for our treatment de- 
cisions, and we know that we desperately need that. I think NCI 
and HHS need to take the suggestions of the National Breast Can- 
cer Coalition and put advocates who are representing and report to 
consumer groups at the table at all the important stages on the 
study sections, on the review committees, on the policy setting 
boards. I think they also need to make a commitment that is 
backed up by policy, by systems, by procedures, to disclose fully 
and quickly the results of clinical research, whether they're good, 
bad, or embarrassing. 

I think if the NCI demonstrates a willingness to take these steps 
instead of doing as it appears they've been doing in the press for 
the past few weeks, just defending themselves, I think they can re- 
gain our trust, and I hope that's the outcome. 

Mr. Brown. Ms. Sigal, can they regain that trust? 

Ms. Sigal. Well, I'd have to say right now I'm probably one of 
the National Cancer Institute's biggest critics, biggest skeptics, but 
I agree with Ms. Pearson. I think it's essential to regain the trust, 
and as I said in my statement, it's not just about breast cancer. It 
could be a study about heart disease, about diabetes or AIDS, but 
it's essential to have trust and credibility in the National Cancer 
Institute, the National Institute of Health, and any government 
sponsored organizations. So, I would hope that could be re-estab- 
lished, but I think it needs to be a different attitude at the Na- 
tional Cancer Institute, and hopefully as a result of this hearing 
and Mr. Chairman's efforts, that maybe the National Cancer Insti- 
tute and National Institute of Health and the efforts of the sub- 
committee members, maybe they'll understand why this is so im- 
portant. Statistically it may not be important to them, but it sure 
is important to me and to thousands of others. 



38 

Mr. Brown. Ms. Visco, in addition to a change of attitude at 
NCI, what steps must they take to regain the trust of people all 
over the country? 

Ms. Visco. Well, I echo Ms. Pearson's support of the National 
Breast Cancer Coalition's demands, including consumer advocates, 
but I want to add something to that. The Coalition's position has 
been from the time we heard about this that we need an independ- 
ent investigation and an independent analysis of the data. We were 
told at first there was a re-analysis. Then we heard that there real- 
ly wasn't a re-analysis and then we're not certain what was actu- 
ally analyzed or re-analyzed. The public trust has eroded com- 
pletely in that area, and we do need an independent analysis of 
what happened. I think that would go a long way toward helping 
restore trust. 

Mr. Brown. Ms. Pearson, what went wrong at Pittsburgh in the 
NCI involving the Poisson matter and the refusals and delays and 
re-analyzing lumpectomy/mastectomy data after the fraud at St. 
Luc's? What actually went wrong? 

Ms. Pearson. Well, I have to just tell you my opinion, and I'm 
someone who's seen Dr. Fisher in action at a few meetings. I've had 
a lot of interaction with NCI officials. As you've heard, Dr. Fisher 
was hailed as an important key researcher in women's breast can- 
cers for many years, and deservedly so. Our founders, in fact, and 
some of our early leaders wrote praises of him in books designed 
for the lay audience. 

I believe that what happened over the 80's was a subtle shift in 
power, to the point where Dr. Fisher, because of his deserved emi- 
nence and respect for his research, began to be more important 
than those who technically funded him and oversaw him. We heard 
the chairman say that it got to the point where he didn't even re- 
turn NCI's phone calls. I knew he wasn't returning mine, and that 
wasn't a real big surprise, but to hear that he wouldn't return Dr. 
Broder's phone calls is shocking to me. 

That's what I think happened, but somehow the NCI's funding 
process keeps going back to the proven producers. Instead of hav- 
ing a process that brings in new blood, fresh ideas, new researchers 
who will have to be accountable to those who fund them, they rely 
too much upon Dr. Fisher's proven track record and let him grow 
in power until he thought he didn't need to be accountable to them 
or his own rules. 

Mr. Brown. What are other scientists and researchers saying 
about what you just said? Are they echoing that? 

Ms. Pearson. Most of the conversations I've had with other sci- 
entists and researchers are very casual and informal. I have heard 
echos of that kind of sense of what's going on, that Fisher was in 
a way untouchable. 

Mr. Brown. Thank you, Mr. Chairman. 

Mr. DiNGELL. The time of the gentleman has expired. The Chair 
is under the discretion that by the rules, the Chair is now going 
to recognize first the gentlewoman from Colorado for questions and 
then the gentlewoman from Maine. Ms. Schroeder? 

Ms. Schroeder. Mr. Chairman, you have been so generous, and 
we really thank you for letting the co-chairs be here. I just want 
to be very brief because I don't want to take the time, but one of 



39 

the questions I have for NCI, and I think that's what you're saying 
when we talk about trust, is they seem to be saying, you as con- 
sumers and you as people who are trying to monitor this, are really 
too stupid to be in at the beginning of this process when they ask- 
ing the questions, framing the questions, and everything. But all 
of a sudden you became very bright at the end of the process. When 
you look at mammograms, they're saying hey, we at NCI can't real- 
ly decide. We'll just have women sit around and read this and de- 
cide whether or not they need it. 

So, it's amazing how bright we come from the beginning. They 
don't want us playing in that sandbox when they're giving out the 
money and framing the questions you want addressed in the re- 
search and monitoring. Then when the research comes out, and ob- 
viously they haven't been monitoring the research too well, they 
kind of say gee, we're confused. Now, if these bright scientists who 
know how to hand out the money become very confused at the end 
and don't know how to hand out the recommendations, we got a 
real problem with attitude. Am I hearing you right? Would trust 
help if you could get in at the beginning? It's a little confusing to 
be told you're so bright at the end after you've been shut out at 
every single level. 

Ms. Visco. Absolutely. I mean, there's no question that would go 
a long way, and probably what would be one thing that would go 
the longest way, in addition to this type of hearing, to restoring 
public confidence is letting consumers in at the very beginning and 
giving us a seat at the table. 

Ms. SCHROEDER. That's right, and I think that goes to the arro- 
gance that we're hearing. I know I think of myself as fairly well 
informed, but I'm really rather enraged when they say sit down 
and read the data and you make the decision whether you would 
need a mammogram between the ages of 40 and 50. Please. I 
mean, are we all to go back to medical school, or what are we sup- 
posed to do? That doesn't make any sense to me. Cindy, I had a 
specific question to ask you that you may or may not want to an- 
swer, and that's on the tamoxifen trial. As you know, there are 
very many, many healthy women enrolled in the current study. As 
we now know late, one of the side effects has been it can cause 
birth defects. 

How confident are you that this trial, the women in this trial, the 
healthy women in this trial, have been notified of that, have been 
told to get birth control, or are there hospitals or doctors who don't 
believe in birth control participating that may not tell them that 
part. Have you looked at that, because another thing about re- 
search is getting people willing to be in the research things. If this 
type of thing goes on, and we'll have that group of women sitting 
at the table saying how could they have known and not told us and 
gotten consent reforms, and we don't want a repeat of the horror 
ofDES? 

Ms. Pearson. That's right. I'm not completely confident that the 
11,000 healthy women who enrolled in the prevention trial before 
this — that it's in right now have been adequately informed about 
the risk of birth defects with tamoxifen. As I mentioned briefly in 
my testimony. Congressman Ted Weiss, before his death, conducted 
an inquiry into the quality of consent that was being given to par- 



40 

ticipants in the trial. His staff reviewed 268 forms from 268 dif- 
ferent centers. Twenty-six percent gave inadequate information 
about the need to use birth control and the t3T)es of birth control 
which were appropriate to be used in the context of this study. 

Upon questioning at the hearing that was chaired by Mr. Payne, 
the response of the prevention staff was well, you know, some of 
the Catholic hospitals are involved, and we can't tell a Catholic 
hospital what to tell women. Now, we agree 100 percent. Catholic 
hospitals are religious institutions, but they don't have the right to 
participate in a trial where women's safety and the safety of un- 
born children depends on their knowledge of the need to use birth 
control and the appropriate kinds. 

Because of Dr. Healy's resistance to that committee's oversight, 
we have no information about whether or not those 26 percent of 
the participating centers have fully complied with the need to give 
women all of the important information. 

Ms. SCHROEDER. Mr. Chairman, that may be something we may 
be able to find out through your committee because if that is still 
going on, I would say stop the trial immediately. I mean, that's out- 
rageous if there are women who are not getting the full consent, 
and let's hope they are. 

Thank you very much, and I thank you, Mr. Chairman. 

Mr. DiNGELL. The Chair thanks the gentlewoman. The Chair is 
going to have the staff look into this matter. The Chair recognizes 
now the gentlewoman from Maine. 

Ms. Snowe. Thank you, Mr. Chairman, and I want to thank all 
of you for your very compelling testimony, and Ms. Sigal, I know 
we're not in your shoes, but we certainly share your agony. In lis- 
tening to your testimony, I couldn't help but think that I would ap- 
proach the issue and the trauma of making such a decision just the 
way in which you did, based on information that my doctor would 
provide and asking questions, and hopefully getting accurate an- 
swers. I just truly regret that this has happened. We just want you 
to know that we share what you are going through. 

Ms. Sigal. Thank you very much. 

Ms. Snowe. What can we do now, immediately, to restore con- 
fidence? I know we've talked about placing a consumer advocate, 
for example, at the table, and I think that's a very important issue, 
and I want to get to that in a moment. What can we do imme- 
diately to restore confidence, because I am concerned about what 
you're going through, and there are thousands of other women who 
are going through the same crisis right now. It's a crisis of con- 
fidence in knowing whether or not you made the right decision. 
Would it be an independent re-analysis as soon as possible? 

Ms. Visco. I feel that would be a very strong statement to the 
American public. If the National Institutes of Health and the Na- 
tional Cancer Institute recognized, not saying they are incapable of 
it, but recognizing the erosion of public trust and because of that, 
stating that they recognize the need for an independent analysis 
and joining with this committee in asking for and in making that 
happen. 

Ms. Snowe. I think that we definitely should insist on it because 
I don't think there's any other way to restore the confidence on this 
issue given the series of terrible errors and misjudgment in ethics. 



41 

I think we have an obligation to do that, and it's something that 
we should do. Have you heard, Ms. Visco, from other women across 
the country as a result of this study? 

Ms. VisCO. Yes. We've heard from quite a few women across the 
country as a result of this study. I will say that most women do 
recognize the existence of other studies, and they have read in the 
papers that the results of the NSABP study will not change, but 
they are still concerned. On an intellectual level, you understand 
that. On an emotional level, you can't accept it. 

Ms. Snowe. Exactly. 

Ms. VisCO. And that's where they are. 

Ms. Snowe. And that's where they are. So, a number of women 
are aware of what has happened based on press accounts? 

Ms. VisCO. Oh, absolutely. 

Ms. Snowe. And obviously the only way they're finding that out 
is as you all did here. 

Ms. Visco. Right. 

Ms. Snowe. In a letter to the co-chairs of the caucus from NIH, 
it talks about the re-analysis that was requested of Dr. Fisher. Is 
it amazing to you that given the fact that they already knew that 
there was a serious breach of ethics and falsification of data, that 
they would go back to the individual who's responsible for conduct- 
ing this clinical study trial to do the re-analysis, and just to as- 
sume that it's going to be done. Obviously he didn't do it. They re- 
quested — they made this request on numerous occasions, and obvi- 
ously he ignored their request, which is just staggering to me that 
somehow the NCI had no ability to enforce this re-analysis on a 
timely basis. 

Should we not make sure that there's an independent coopera- 
tion any time that there is a problem and question about the accu- 
racy of the data? 

Ms. VisCO. Yes, absolutely, I agree, especially after this episode. 
Perhaps if this episode had been handled differently, we'd have a 
different answer to that question, but I think in light of recent 
events, absolutely we'd need it, as part of the process. 

Ms. Snowe. Because I am just amazed that Dr. Fisher would, for 
whatever reasons, refuse on numerous occasions — I mean, there 
were so many time lags between the time they identified the dis- 
crepancies and the time in which they received any official report, 
and then it was sort of tucking this all into the Federal Register 
in a very brief statement in hopes that no one is going to identify 
it or catch it. I mean, that was obviously the intent because every- 
body knew then that something was seriously wrong here that they 
wanted to go unnoticed, but they wanted to be on record that some- 
how they had identified this problem and they made public recogni- 
tion of that problem. 

What are the next steps that we should take? You were mention- 
ing having an advocate in the process. Recently, I co-signed a letter 
with Representative Towns asking for a consumer advocate on the 
consensus panel for ovarian cancer. Is that what you're talking 
about as well? 

Ms. Visco. Yes. 

Ms. Snowe. And at the beginning stages of the process, should 
that be by statute or should it be by regulation? 



42 

Ms. ViSCO. I'm not certain it can be by statute. I believe it should 
be policy of the National Institutes of Health and the National 
Cancer Institute, that when Federal money is being spent — we're 
working on getting this done on a private institutional level also, 
but you can be instrumental in helping us make it happen on a 
public level, that consumer representation belongs at the table 
study sections, data monitoring committees, oversight committees, 
not just advisory boards. 

Ms. Snowe. Well, the reason why I asked whether it should be 
done by regulation or statute is because we had a problem with the 
fact that women were excluded systematically from clinical study 
trials, and actually, NIH was violating its own policy. It was not 
enforcing its own policy. So, I have mixed feelings about whether 
or not that should be done by regulation or statute so that we have 
the assurance that it is being done and that we don't have to go 
through the process of determining whether or not it's being en- 
forced or not, that we know by statute they are required, and obvi- 
ously we understand that there could be problems there, too, but 
it's obviously going to be much harder for them to contravene pub- 
lic law. 

Mr. Brown. Absolutely. When you're handing out money on a 
Congressional level and on a National Cancer Institute level, you 
certainly have the power to make as a condition to being eligible 
for that money, the fact that a consumer representative must be in- 
volved. 

Ms. Snowe. Well, thank you all very much. 

Mr. DiNGELL. The time of the gentlewoman has concluded. The 
Chair would like to express my personal thanks to each of you for 
your valuable assistance to the committee. Your help has been of 
great importance to us in developing our record. The Chair advises 
that the questions that you have raised, including the chronology 
of events, will be brought to the attention of witnesses from the 
NIH. The Chair excuses the panel members with our gratitude. We 
also wish you, Ms. Sigal, great good fortune. 

Ms. Sigal. Thank you very much, Mr. Chairman. 

Mr. DiNGELL. And you also, Ms. Visco. I want to extend in addi- 
tion to that my personal thanks. Thank you very much, ladies. 

Ms. SiGAL. Thank you, and thank you, Mr. Schaefer. 

Mr. DiNGELL. The Chair announces the next panel. The panel 
will be composed of Dr. Harold Varmus, Director of the National 
Institutes of Health, and Dr. Samuel Broder, Director, National 
Cancer Institute. They will be accompanied by Dr. Bruce Chabner, 
Director, Division of Cancer Treatment, Dr. Michael A. Friedman, 
Associate Director, Cancer Therapy Evaluation Program, and Dr. 
Lyle W. Bivens, Director, Office of Research Integrity. Gentlemen, 
thank you for being with us today. We express the thanks of the 
committee to you for your presence. The Chair advises that, as you 
know, the practices of the committee are that all witnesses who 
testify before the committee testify under oath. Gentlemen, do any 
of you have any objection to testifying under oath? Very well, the 
Chair advises that if you do testify under oath, it is of course your 
right to be advised by counsel in the course of your appearance. Do 
any of you desire to be advised by counsel as you testify under 
oath? 



43 

[No response.] 

Mr. DiNGELL. Very well. The Chair advises that copies of the 
rules of the subcommittee, rules of the committee, rules of the 
House are there at the witness table before you in the red and blue 
books to inform you of your rights and limitations on the power of 
the committee as you testify before us today. Grentlemen, if you 
have no objection then to testifying under oath, would you please 
each rise and raise your right hand? 

[Panel sworn.] 

Mr. DiNGELL. Gentlemen, you can consider yourselves to be 
under oath. Gentlemen, we will recognize you beginning with Dr. 
Varmus, then Dr. Broder, and then by Dr. Bivens, and then by the 
others as you might find it necessary or might suit the needs of the 
hearing. Doctor, you are recognized. 

TESTIMONY OF HAROLD VARMUS, DIRECTOR, NATIONAL IN- 
STITUTES OF HEALTH; SAMUEL BRODER, DIRECTOR, NA- 
TIONAL CANCER INSTITUTE, ACCOMPANIED BY BRUCE 
CHABNER, DIRECTOR, DIVISION OF CANCER TREATMENT, 
MICHAEL A. FRIEDMAN, ASSOCIATE DIRECTOR, CANCER 
THERAPY EVALUATION PROGRAM; AND LYLE W. BIVENS, DI- 
RECTOR, OFFICE OF RESEARCH INTEGRITY, DEPARTMENT 
OF HEALTH AND HUMAN SERVICES 

Mr. Varmus. Mr. Chairman, thank you. Before I proceed with 
my prepared statement, I would like to say a word or two about 
the transition in mood that will accompany the development of this 
panel. We've just experienced an emotionally trying discussion by 
three women who have all been affected by or are involved in the 
ramifications of the breast cancer studies that we've talked about. 
Ms. Sigal's testimony was particularly impassioned and moving. 
We're going to now move into a more analytic mode to discuss the 
process by which the current events unfolded. 

As we do that, I think it's important to realize that although we 
are five men at this table responsible for some of the events that 
we want to explore and for their oversight, it's important to re- 
member that we, too, have a deep, passionate involvement in these 
issues. In my own case, my mother and my grandmother died of 
breast cancer. I've devoted most of my research career to the pur- 
suit of an understanding of breast cancer. I have dear friends who 
are making the kinds of decisions Ms. Sigal and Ms. Viscoe were 
describing. I feel a deep responsibility for the research that leads 
to the ability of women and their physicians to make such deci- 
sions. 

We have to recognize as well that as the previous panel indi- 
cated, we're talking not just about gender issues and breast cancer, 
we're talking about all of the research that's affected by NIH. We're 
talking about events that could have affected a study of prostate 
cancer or lung cancer or other diseases that are rare or common. 
We take these responsibilities very seriously, and as we engage in 
this discussion and proceed to a detailed analysis of how we've re- 
sponded to the events that you've heard described, that although 
we analyze these, the passion that you've heard in the previous 
panel is shared my many of us. 



44 

That being said, Mr. Chairman, we welcome the chance to ap- 
pear before you today to discuss the disturbing events that have 
been well described here in the previous panel and by yourself and 
in the press that involve the clinical studies of breast cancer car- 
ried out over many years by the NSABP under the sponsorship of 
the NCI. Before I introduce Dr. Broder, the Director of the NCI, 
who will provide a detailed account of these events and his efforts 
to respond to them, and before we introduce his colleagues and 
Lyle Bivens from the Office of Research Integrity, I would like to 
outline what I hope we can accomplish during this panel. 

First and foremost, we can reassure women, including Ms. Sigal, 
who have chosen breast sparing surgery for breast cancer, that 
they have made a decision that is strongly and unequivocally sup- 
ported by current scientific evidence, evidence provided by both the 
NSABP and by many other groups. That analysis has been inde- 
pendently analyzed by the EMMES Corporation, as Dr. Broder will 
review. We are prepared to submit all the data to independent au- 
diting if that proves to be necessary. 

Second, we can summarize the central facts in a complex story 
that includes a long history of path breaking clinical science by the 
NSABP, leading to many of the improvements in the treatment of 
this disease, which regrettably we still incompletely understand. It 
also includes a well documented episode of flagrant fraud that 
you've heard discussed at one of its more than 400 participating in- 
stitutions, and the facts include inadequate and slow administra- 
tive responses and insufficient auditing procedures by arms of the 
government and by the NSABP. 

Third, we hope we can describe corrective measures that have 
been taken in recent weeks by the NCI to improve the administra- 
tive functions of the NSABP and to strengthen oversight by the 
NCI. Such measures should increase the likelihood that fraudulent 
acts occurring in the conduct of clinical studies will be promptly de- 
tected and reported in the future. In this way, we believe that the 
lessons learned from the current incident can enhance the conduct 
of medical science supported by the NCI and by the other compo- 
nents of the NIH. 

It's important to recognize at the outset that remarkable 
progress has been achieved in the practice of medicine and surgery 
in the past few decades. Consider just three prominent examples: 
The cure of acute leukemia in children by chemotherapy, the reduc- 
tion in mortality by control of high blood pressure, the prevention 
of blindness by laser surgery in diabetic patients. In each of these 
three instances and in many other examples, including the breast 
cancer therapies under discussion today, progress can be attributed 
to the formation of large cooperative, multi-institutional and often 
multinational groups. 

The well documented successors of such groups require meticu- 
lous planning, coordination, and oversight to insure that protocols 
are uniformly followed in an often varied and far flung group of in- 
stitutions. Occasionally, as in the case of the NSABP, such proce- 
dures may fail, and the malfeasance of an investigator may go un- 
detected and uncorrected. The hard-earned prestige of productive 
groups such as the NSABP may even be among the factors that 
contribute to the deficiencies of monitoring and reporting. 



45 

Our discussion today will properly focus not on the act of fraud 
itself but on the methods for detecting, investigating and publiciz- 
ing fraud and for evaluating its consequences. The questions we 
will address are especially contentious and complex when the sus- 
pected fraud might affect the outcome of clinical studies and hence 
might alter the procedures used in medical practice. Under such 
circumstances, there are profound tensions between the public's 
need for access to information that could affect people's health and 
the right of someone accused of fraud to be protected by due proc- 
ess. 

The public does not want to be frightened by false accusations. 
Likewise, the public does not want to be deprived of knowing the 
consequences of accurate ones. For these reasons and others, as the 
previous panel has stressed, the prompt resolution of allegations of 
fraud and the rapid dissemination of findings of fraud in clinical 
research are matters of great importance to us, and they deserve 
our attention today and in the future. 

It's my hope that this hearing can provide an opportunity to un- 
derstand what has happened in the current instance and to ac- 
knowledge the dedication and integrity of the vast majority of clini- 
cal researchers whose valuable efforts may be jeopardized by the 
recent events. It's an opportunity, most importantly, to reflect upon 
ways we could learn from this experience to improve what we do 
to advance the Nation's health and to insure that patients such as 
those you've heard in the first panel receive the benefits of our ef- 
forts as caretakers of the Nation's health. 

At this point, Mr. Chairman, I'd like to introduce Dr. Sam 
Broder, the Director of the NCI, who will give you an account of 
his institution's response to these events. 

STATEMENT OF SAMUEL BRODER 

Mr. Broder. Good morning, Mr. Chairman. Good morning, 
Madam Chairwoman and members of the subcommittee. I am Dr. 
Samuel Broder, Director of the National Cancer Institute, the NCI. 
Accompanying me today on your left are Dr. Bruce Chabner, Direc- 
tor of the Division of Cancer Treatment of the National Cancer In- 
stitute and Dr. Mike Friedman, the Associate Director of the Can- 
cer Therapy Evaluation Program. We are very pleased to have this 
opportunity to follow up on Dr. Varmus's remarks. 

I would like to discuss NCI's recent actions following the Office 
of Research Integrity findings by fraud on the part of a surgeon at 
L'opital St. Luc in Montreal, and the steps taken by NCI to correct 
this situation. The oversight and monitoring of clinical trials is 
vital to insure that research advances are based on sound and ac- 
curate data. Honest errors can and do occur, and will always occur 
whenever human beings are involved in any process. However, 
fraud is different and cannot be tolerated. We have a responsibility 
to all cancer patients, physicians, and scientists to validate the in- 
tegrity of our clinical research. 

There is in this country widespread use of breast sparing proce- 
dures as an alternative to mastectomy. I am not that old, or I don't 
think I'm that old, but within my medical school training, there 
was an era in which women would be informed that they had a 
biopsiable lesion, would be given a general anesthetic, would have 



46 

no option except to go under the surgery, and to not know whether 
they would awake with a benign biopsy or what was then a radical 
mastectomy. A medical student in my era who challenged this 
practice of medicine in that time would possibly face problems with 
his or her career development. 

Today, I assure you, Mr. Chairman, members of the subcommit- 
tee, the general public, and physicians, that breast sparing surgery 
remains an appropriate and safe procedure. I do not come here be- 
fore you as merely a scientist or government administrator per se, 
but as someone who knows how breast cancer can devastate 
women and entire families at first hand. No woman who has relied 
on the results of modem clinical trials to select a breast sparing 
procedure following a diagnosis of invasive breast cancer should 
feel that she has made the wrong choice based on the topics under 
discussion today. 

The principal of breast sparing surgery is based on a set of at 
least six international studies and a couple that are not yet pub- 
lished, which provide the clear basis for current medical practice. 
The issue of ipsilateral breast cancer that we've heard, that is a re- 
currence in the same breast, may, as the facts were presented in 
earlier testimony, may possibly be Dr. Fisher's view, but I wish to 
assure the committee that it is more assuredly not my view, and 
I believe I have made that point clear to the staffers who have been 
kind enough to speak to me on this point. We certainly consider 
ipsilateral breast cancer recurrence to be a valid indicator of the 
process, and in fact, lumpectomy and radiation compares quite fa- 
vorably in this connection to other available therapeutic ap- 
proaches. 

Before discussing the details of this case, Mr. Chairman, I be- 
lieve that documents summarizing the NCI's own statistical analy- 
sis done through a contractor who is an agent of the National Can- 
cer Institute, that those documents dealing with breast sparing 
surgery and other treatment approaches have been provided to the 
committee, and we are also making these reports, the reports that 
pull out the St. Luc's data widely available, employing computer 
networks that will make the results as widely available as possible. 

In addition, perhaps it is of some interest to the committee that 
we had outside reviewers and tried to involve statistical experts 
with a known consumer advocacy position. Dr. Kay Dickerson has 
reviewed the NCI reports of the statistics excluding the fraudulent 
data, and she writes, and I'm quoting, "The ansdyses were clearly 
presented, and seemed to be properly done." 

Before discussing the details of this case, perhaps I could briefly 
comment on the NSABP. The group has existed for over 35 years 
as a leading clinical trials component in the breast cancer field. It 
consists of several thousand doctors at more than 400 sites and is 
one of 9 clinical trials cooperative groups that conduct large scale 
clinical studies supported by the National Cancer Institute in the 
United States and Canada. The NSABP's founder. Dr. Bernard 
Fisher, constructed the group as a broad based organization of com- 
munity and academic surgical oncologists whose findings would re- 
flect the ability, the capabilities and the imperfections of practicing 
physicians in the real world to employ innovative therapies. The 
NSABP has pioneered and established many principles of breast 



47 

cancer treatment that have, in fact, been repUcated by other orga- 
nizations and groups. 

Please let me say something that I hope is not misunderstood in 
testimony before this committee. Today, Dr. Fisher's prior record of 
scientific preeminence, indeed even the fact that he as a surgeon 
played a special role in liberating women from having to undergo 
mastectomies, all of this is irrelevant to reviewing what went 
wrong and how we should do better. 

In June 1990, the NSABP staff detected inconsistent data in the 
record of a patient previously entered on an NSABP trial in Mon- 
treal. This was at L'opital St. Luc. In two site visits of the ensuing 
7 months, NSABP became convinced that fraud had taken place, 
and in February 1991, notified the NCI for the first time. The NCI 
notified the Office of Scientific Integrity, the precursor of the Office 
of Research Integrity, which will be represented by my colleague. 
Dr. Lyle Bevins, and also notified the Food and Drug Administra- 
tion and the Office of Protection from Research Risks. 

Government audits were organized and carried out, and they 
quickly confirmed the findings of fraud. To be candid, this was not 
a difficult investigation from my point of view. The investigator, 
Dr. Roger Poisson, admitted the forgery and falsification of dates 
and other fabrications in the records of a small number of patients 
that he entered into NCAB. I use the small number as his point 
of view, not my point of view. His later justifications for these acts 
are incomprehensible to us. They remain incomprehensible to us to 
this day. 

OSI — now ORI — undertook a detailed investigation. During such 
an investigation, ORI generally prefers to have an embargo on pub- 
lic discussions and disclosures. However, Dr. Fisher was asked by 
both NCI and ORI, orally and in writing, to perform a re-analysis 
and prepare a publication of the trials affected by the fraud. It is 
my personal view that had he listened to us or had we been able 
to force him to listen to us, this hearing would be unnecessary. As 
early as July, 1991, an SABP assured NCI staff that outcomes were 
unchanged. In March, 1992, Dr. Fisher presented his re-analyzed 
data to medical and statistical of OSI, NCI and NIH at a presen- 
tation in Bethesda, and concluded there were no changes in the 
major end points of the affected studies. 

NSABP was asked to prepare a manuscript to publish their re- 
analysis at the conclusion of the misconduct investigation. Neither 
I nor NCI felt there were any urgent public health hazards or 
changes in medical practice that would warrant breaking the em- 
bargo which ended in April, 1993 with the release of the final ORI 
report. We are re-assessing the algorithm by which we decide 
whether an embargo needs to be broken, and this particular kind 
of faith to an embargo or an adherence to an embargo is unlikely 
to ever be repeated, at least from my point of vie\/. 

The report's findings of fraud were summarized in a number of 
formats which are not satisfactory and include, of course, the NIH 
guide in June of 1993. Each issue of the guide is mailed to over 
36,000 subscribers and is available on computer networks, but 
there can be no doubt that ORI and NCI put Dr. Fisher on notice 
to publish a re-analysis. It is clear, however, that the methods used 
to announce the results of this misconduct were not adequate, and 



48 

we will approach the dissemination of such results working closely 
with ORI in a more active and visible way in the future. 

In February 1994, the NSABP sent NCI a written summary of 
their re-analysis of clinical trials, including the breast sparing pro- 
cedure study. Shortly thereafter in March of 1994, accounts in the 
media generated concerns about the delays in publication and 
raised the possibility that the practice of medicine might be based 
on faulty conclusions. I believe it is Dr. Fisher's assertion or the as- 
sertion of his staff that he was preparing to publicize these results 
in June of 1994. 

There is cause for reflection and review of actions regarding what 
the NCAB and NCI did in responding to these issues. First, let us 
look at NSABP. The NSABP failed to publish its re-analysis, in- 
form its membership of the incident, reassure the public, notify sci- 
entific journal editors and other grant supported organizations of 
the fabrication, public accurate papers that clearly disclose what 
Dr. Poisson did, and in a larger sense, adhere to NCI's guidelines 
for management of the group's operation office and quality assur- 
ance functions. The NSABP did not respond to constructive criti- 
cism by NCI staff. 

Unfortunately, the problems of the NSABP have continued. In an 
NCI on site review of their operations in Pittsburgh toward the end 
of March of 1994, NCI staff found additional evidence of deficient 
auditing and reporting practices. The staff found that the NSABP 
had failed to conduct required audits of its treatment studies in a 
timely fashion and had failed to transmit to NCI reports of certain 
audits of its treatment and prevention trials as required. 

In the files, we found a report of an additional episode of possible 
data manipulation involving a patient on a different study at yet 
a different hospital in Montreal. This finding was initially discov- 
ered by the NSABP in September of 1993 but had not been re- 
ported to NCI and was, as far as we can tell, being handled as a 
routine matter subject to an additional follow-up site visit. The NCI 
staff initiated an emergency site visit to the new institution in 
question, confirmed a suspicious alteration in an X-ray report, and 
immediately notified ORI, which is now investigating the matter. 

That latest irregularity appears to involve a clinical study, the 
tamoxifen breast cancer prevention trial that is not completed and 
therefore is unpublished. I have been informed by ORI that the 
breast cancer treatment results were not tainted or do not appear 
to be tainted by this latest irregularity in Montreal. 

The NSABP had no explanation for its failure to comply with re- 
quests to update and report its audit findings nor for its delay in 
informing its membership of these problems and promptly submit- 
ting its re-analysis for publication. We believe that the primary re- 
sponsibility for correcting fraudulent work in the literature lies 
with clinician authors of the original publication. However, as I 
mentioned earlier, the NCI clearly has responsibilities as well. 

Now, let us look at NCI. Despite the requests by NCI and ORI 
for the group to re-analyze and publish the data and to bring oper- 
ations into compliance, we failed to compel these actions forcefully 
and in a timely way. We also did not adequately overcome our re- 
luctance to demand that an independent investigator, who himself 
was not a respondent in a misconduct case, turn over his data files 



49 

in their entirety to have other re-analyze the results. We are trying 
to learn from this experience to ensure that our response of epi- 
sodes of fraud in clinical trials is prompt and effective. Thus, re- 
cently NCI personnel have taken possession of the computer data 
files, analyzed and disseminated the results, initiated a govern- 
ment run, on- site audit of clinical research conducted by the group 
and certain other groups. 

In addition, in doing so, NCI has clearly confirmed the principle 
that the granting agency can and will demand, distribute and dis- 
close a grantee's data in response to pressing public health needs. 
Fraud will by definition always constitute such a need. We will not 
tolerate explanations that the data belonged to the grantee. We 
will never again hesitate to exercise this authority whenever nec- 
essary. 

Our staff also failed to mobilize after warning signs of delay in 
clearing up the scientific literature and repairing inadequate com- 
pliance with auditing requirements. This may have occurred be- 
cause of several factors taken alone or in combination, which I be- 
lieve the committee will wish to note, and these may be: NSABP's 
proud reputation; the visible status of Dr. Fisher in the scientific 
community and also as a member of our presidentially appointed 
National Cancer Advisory Board; a self-consciousness in asserting 
authority over an independent researcher; or a mistaken belief that 
somehow the lapses by a senior research group were temporary. 
We are very sorry that this happened, and we assure you that such 
errors will not happen again. We have created a new unit, the Clin- 
ical Trials Monitoring Branch, to monitor compliance in our cooper- 
ative groups, which will implement our rules without fear or favor. 
We will take swift and uninhibited action in the event of a lack of 
compliance. A prior record of accomplishments of any level will not 
be used as a defense against adhering to our regulations. New pro- 
cedures are in place to report and track audits, and we are initiat- 
ing a system of NCI-directed site visits to validate the cooperative 
group audit findings with sites selected at random. 

Also, we are reviewing our computer security and password-only 
entry procedures for the various clinical trials in which a central 
laboratory runs lab tests on auto analyzers with transfers to a com- 
puter database in order to avoid the possibility of the scientific mis- 
conduct through a computer hacker could affect centralized com- 
puter operations, and we want to address this issue. 

We are developing a new internal NCI operations manual for sit- 
uations involving fraud and scientific misconduct. This manual will 
be in place within days. The messages also include automatic noti- 
fication of journals where falsified data had been publish, and in 
forming other financial sponsors of projects affected by scientific 
misconduct. Any administrator at the NCI can initiate an auto- 
matic series of steps with checklists, enabling NCI as a whole to 
act decisively and resolutely while at the same time protecting the 
legitimate interests of all parties. These measures will protect the 
public health and provide for the recovery of Federal funds. The 
new branch has begin to review policies and procedures for mon- 
itoring trials and plans to revise and amplify the guidelines for 
every aspect of the process. Dr. Michelle Christian, who is the act- 
ing chief of this new branch, is in the hearing room today. 



50 

A number of new policies are in force. All site visits will include 
an individual from outside the cooperative group conducting the 
audit. All audits will be on site. In addition to any other regular 
audit and oversight procedures, the NCI, using Federal employees 
and its own contractors, will conduct special audits of cooperative 
groups, sites selected at random, to verify the accuracy of coopera- 
tive group audits. This will impose an external element of unpre- 
dictability and surprise in the auditing and oversight functions. 

Accrual of new patients will automatically be suspended for insti- 
tutions that have not had a site visit with regular 3 year cycles. 
Audit schedules and reports will be computerized. The NCI is ex- 
ploring a common on-line computer system to be shared by the 
groups. The peer review site visit process will be used not only for 
scientific review, but also to help identify problems in protocol com- 
pliance and quality assurance. 

The branch will issue a quarterly report to the Executive Com- 
mittee of the NCI and the NCI components that are doing clinical 
trials. Grantees that fail to meet our requirements for accuracy, re- 
liability and time limits will be suspended. 

We appreciate the need to improve the flow of information on 
drug toxicities between NCI and the cooperative groups. The early 
detection of side effects poses special challenges in large commu- 
nity-based clinical trials. As a general rule, in the future, we will 
try to have the National Cancer Institute in effect actually hold the 
investigation of a new drug application, what is called the INDA, 
for studies that we sponsor. 

Except under compelling circumstances, I must tell you candidly 
that this is not always an investigator's or a pharmaceutical com- 
pany's first choice. But holding the INDA enhances certain types 
of reporting and additional oversight issues beyond grand manage- 
ment and larger clinical trials. For example, in the tamoxifen ran- 
domized trials, NCI did not hold the IND, and therefore would not 
necessarily have been the very first in line to receive data on new 
cases of endometrial cancer or other potential complications from 
tamoxifen. We will try to communicate in novel ways directly with 

Eatients on clinical trials. For example, we will tests on a pilot 
asis an electronic patient information bulletin board to provide pa- 
tients with the latest communications specific to a given trial. This 
is to supplement, not replace, our other ways of communicating 
with patients and consumers, including our cancer information 
service, cancer facts, and physician data query systems. 

What about the current status of the NSABP? The NSABP is an 
important resource for conducting large scale randomized clinical 
trials in this country. A number of changes have now been made 
to strengthen the operation of NSABP. The NCI instructed the 
granting institution to change the principal investigator, and Dr. 
Fisher, the leader of the group since its inception, has stepped 
aside. I do not recall a similar request ever having been made with- 
in my historical knowledge of tne NCI on a major figure of Dr. 
Fisher's status. Dr. Ronald Herberman, the director of the Cancer 
Center at the University of Pittsburgh and a widely respected sci- 
entist, has assumed this job on an interim basis. An interim execu- 
tive officer has been appointed to oversee daily operations. An over- 
sight committee composed of experienced oncologists, breast cancer 



51 

specialists, and a consumer advocate has been appointed, and the 
membership of the group has received information regarding the 
fraud and operational deficiencies of the group. 

It is my personal belief that Dr. Fisher pioneered and established 
this group prior to even the founding of the national cancer pro- 
gram in its current iteration, and I believe that is a fact that has 
partially influenced the relationship of who reports to whom. The 
group has been placed on probation, and it has until the end of 
June 1994 to initiate programs to bring its auditing and reporting 
procedures into compliance. Accrual to clinical trials has been tem- 
porarily suspended until these deficiencies are correct and a quality 
auditing system is in place. 

The terms of awards of the NSABP grant have been modified to 
require immediate republication of any trials affected by fraud. 
Similar grant conditions are being implemented for all of NCI's 
clinical trials' cooperative groups. Our own NCI run audits of var- 
ious hospitals and academic centers affiliated with this group con- 
tinues. We are attempted to recover funds expended at the fraudu- 
lent data site. We consider the entire data set from St. Luc's Hos- 
pital to be a total loss to the American taxpayers. 

We have learned a great deal about the process and pitfalls of 
dealing with scientific misconduct. We clearly understand the prin- 
ciple that we cannot allow a grantee's formidable reputation, his- 
tory of prior accomplishments or service in science to stand in the 
way of prompt, corrective action and oversight. Prior achievements 
do not render individuals immune to what we must get from them. 
We cannot and will not ever defer or appear to defer to the time- 
table of a grantee of whatever preeminence in reporting fraud and 
fabrication to the public. We have taken steps to make this the ex- 
plicit policy of the National Cancer Institute. 

We hope that the measures outlined above will bolster at least 
in part the confidence of the Congress, the public, and the medical 
community in our clinical trials program and will strengthen the 
operation of our productive clinical trials groups, most of whose 
members honor careful and honest science and are devastated by 
episodes of fraud, as are we at the National Cancer Institute. 

I thank the Chair for his very generous permission to allow me 
to go over my time limit. Thank you for providing me this oppor- 
tunity to testify and present the facts about NCI's role. I would be 
pleased to answer any questions. 

Mr. DiNGELL. Thank you, Dr. Broder, for a very helpful state- 
ment. Dr. Bivens? 

TESTIMONY OF LYLE W. BIVENS 

Mr. BrvENS. Mr. Chairman, I'm pleased to have the opportunity 
to review for the subcommittee the response of the Office of Re- 
search Integrity to the allegations of scientific misconduct at St. 
Luc Hospital. Although the Office of Research Integrity was estab- 
lished in June 1992 and was therefore not involved in the case 
until that time, I'll attempt to provide briefly the general back- 
ground of the circumstances leading to ORI's specific role in the 
case. I'll then describe the nature of our investigation and subse- 
quent activities. I beg your indulgence for some duplication here. 



52 

but I think it's important for the record for me to cover background 
material leading up to our involvement. 

In June 1990, the staff of the National Surgical Adjuvant Breast 
and Bowel Project, a large collaborative clinical trial supported by 
the National Cancer Institute, discovered discrepancies in two re- 
ports of breast cancer surgery involving an individual who was the 
study subject. This woman was part of the study group from St. 
Luc Hospital in Montreal, one of the many hospitals participating 
in the study. 

In September 1990, an audit by NSABP staff identified problems 
with both medical records and documents relating to the informed 
consent. Dr. Poisson, the principal investigator for the St. Luc com- 
ponent of the study, was informed about these problems in Decem- 
ber. 

Early in 1991, NSABP reviewed over 100 cases and found five 
discrepancies in this group. These discrepancies had the effect of 
making five patients eligible for inclusion in the study who would 
have not been eligible had the protocol been followed properly. 

On February 6, 1991, Dr. Bernard Fisher, the principal inves- 
tigator for the overall project, suspended accrual of new patients 
from St. Luc and asked for an explanation of the data discrep- 
ancies. Dr. Poisson admitted to alteration of records in a few in- 
stances but not to any wrongdoing. On February 14, NCI received 
notification from Dr. Fisher that irregularities have been found in 
the data from St. Luc. 

On February 22, some 8 months after the initial problems were 
identified, NCI staff met with Dr. Fisher at Pittsburgh to discuss 
the findings. At the conclusion of the meeting, NIH's office of sci- 
entific integrity was notified by NCI of apparent fabrication and 
falsification of the data from St. Luc. NCI also notified the Office 
for Protection from Research Risks and the FDA because of in- 
formed consent problems and the fact that some of the questioned 
data had been filed under an investigational new drug application 
for tamoxifen. 

OSI immediately decided to conduct a direct investigation due to 
the importance of the NSABP studies in changing standards for 
medical practice. 

On March 1, 1991, OSI opened a formal investigation and di- 
rected that all records for the patients be secured and maintained 
under proper custody. OSI developed a plan to audit a sample of 
cases entered on NSABP studies from St. Luc hospital. This was 
a complex task. Three charts were maintained for each case, in pa- 
tient records, outpatient records and research records and the med- 
ical records were in French. 

An OSI team along with staff from NCI, NSABP and FDA, made 
an initial visit to St. Luc Hospital on April 22 to 24, 1992, to re- 
view records and interview St. Luc staff. The review team found 
that 5 to 10 percent of the cases in the sample contained what ap- 
peared to be data discrepancies. OSI decided to audit records of all 
1,504 patients from St. Luke hospital rather than a sample of 
cases, as was the original intent. Many of the discrepancies were 
altered dates of procedures or tests that made patients eligible for 
inclusion in the research protocol, whereas the correct information 
would have made them ineligible. 



53 

Subsequent review of records was conducted on June 11 to 14 
and 25 to 28, and on August 26 to 18. Beginning in September, OSI 
and NCI evaluated all the information and documents collected 
during the investigation and compiled a database of falsification 
and fabrication charges. On February 19 to 21, 1992, OSI revisited 
St. Luc hospital to confirm details on questionable cases. 

An overriding concern from the beginning of this investigation 
was whether the scientific basis for breast cancer treatment strate- 
gies was invalidated bv the falsified and fabricated records. NCI 
and OSI requested early on that NSABP determine whether clini- 
cal outcomes were changed. It was requested that NSABP revise 
the research records with the questionable data excluded. Oral con- 
firmation was provided by NSABP in July, 1991, that outcomes 
had not changed. At the request of OSI, NSABP presented the re- 
analyzed data to OSI, NCI, and NIH staff in March 1992. NCI and 
OSI statisticians were aware that this was a preliminary analysis, 
and then NSABP was asked to prepare a manuscript to publish the 
re-analysis by the time the misconduct investigation was com- 
pleted. 

Because the preliminary re-analysis indicated that the basic find- 
ings of the study remained valid, OSI concluded that no clinical 
alert or notice was necessary and, consistent with existing policy, 
no public information on the case was provided while the investiga- 
tion was underway. 

On April 8, 1992, the OSI again visited St. Luc Hospital, this 
time with two outside experts to review the cases. From June 
through November, 1992, the newly established Office of Research 
Integrity examined all of the evidence, prepared a detailed written 
report for each case, solicited responses from Dr. Poisson on each 
case and prepared an overall report which was sent pursuant to 
our standard procedures to Dr. Poisson for comment in November 
1992. 

Although Dr. Poisson admitted to altering records early in our 
investigation, his admissions were very limited and provided in a 
piecemeal fashion, only after further discrepancies were noted as 
our audits proceed. It was essential for investigation to uncover the 
full extent of the misconduct, especially because of the potential 
clinical significance of the data in question. Therefore, we did not 
stop our process as soon as Dr. Poisson made a very limited admis- 
sion. 

On January 5, 1993, the NCI sent a letter to Dr. Fisher urging 
finalization of the manuscript on reanalysis of the published stud- 
ies. On February 19, 1993, ORI notified Dr. Poisson of our findings 
and proposed actions, including his debarment from receiving Fed- 
eral grant or contract funds for years. Copies of our notice and final 
report were also sent to NIH, St. Luc Hospital, and the University 
of Pittsburgh. At the same time, I wrote a memorandum to the di- 
rector of NCI summarizing OKI's recommendations for publication 
of the reanalyzed studies. 

Under a newly establish hearing opportunity for scientists 
charged with misconduct. Mr. Poisson had 30 days after initial no- 
tification to request a hearing. He did not do so, and ORI made its 
findings final effective March 29 and notified Dr. Poisson of this on 
April 16. 



54 

At about the time that this case was being concluded, ORI had 
decided that it was in the public interest to provide wide notice of 
our findings of scientific misconduct. This was a major departure 
from the practice of our predecessor organizations in which infor- 
mation on cases of confirmed scientific misconduct was released 
only in response to requests under the Freedom of Information Act. 
A Federal Register notice was prepared, summarizing all fourteen 
of the cases of scientific misconduct that had been closed since ORI 
was established in June 1992, and this notice was published on 
June 21, 1993. 

We also published the identical information in the NIH guide to 
grants and contracts, a weekly publication sent to more than 
30,000 institutions, hospitals, ana organizations concerned with 
NIH research. We described the St. Luc case in the April issue of 
the ORI newsletter, which is sent to all institutions receiving pub- 
lic health service research funds, as well as to a large number of 
professional and scientific organizations. 

In addition, we issued a brief press release announcing findings 
in the 14 completed investigations, including the St. Luc investiga- 
tion. In addition to these releases, we provided the full report of 
our investigation of St. Luc's hospital in response to six requests 
under the Freedom of Information Act. Two of these requests were 
from the media and four were from institutions or individuals. 

If, during the course of the investigation, ORI had identified any 
immediate threat to the scientific basis for breast cancer treatment 
strategies, we would have taken immediate steps to alert the public 
and those making treatment decisions based on those studies. 
There was no such indication, so we adhered to our standard proce- 
dure, waiting until the investigation was complete before releasing 
information. 

I believe that ORI properly discharged its responsibilities in this 
case, conducting a thorough and objective investigation and dis- 
seminating the findings widely. However, it is also clear from this 
case that we can improve our effectiveness when we find mis- 
conduct in science. Consequently, we have initiated a number of 
changes. First, we've instituted a policy to routinely and directly 
notify journal editors of our findings whenever it appears that cor- 
rections or retractions of the scientific literature are warranted. 

Second, we're working more closely with the PHS funding agen- 
cies in following up our recommendations when misconduct is 
found or even in cases where there is no misconduct but some re- 
medial action seems required. We established a process whereby 
we request a report within 45 days of our notifying a PS funding 
agency, and that certain actions will be taken. This will assure that 
ORI recommendations are flagged for the PHS, and acted upon in 
a more effective and coordinated fashion by the Department. 

Third, and perhaps most important, based upon the testimony I 
heard this morning, I believe that ORI should provide a higher vis- 
ibility press release at the conclusion of any investigation finding 
misconduct in a clinical trial whether the findings of the trial are 
invalidated or not. If we had done this in the present case, it might 
have allayed unnecessary concerns about the validity of the find- 
ings of the study. This was our first finding of scientific misconduct 
in a clinical trial, and our overriding concern was whether or not 



55 

there was any immediate need to provide the medical community 
with some sort of clinical alert that the scientific basis for a treat- 
ment strategy might be called into question. However, it is clear 
that we also need to attend to the concerns and perceptions of 
those persons most directly affected by the clinical research in 
question. 

I'll be pleased to respond to any questions. 

Mr. DiNGELL. Thank you very much, Doctor, for your very wel- 
come statement. Dr. Broder, you were speaking on behalf of the 
NCI. Dr. Varmus, you're here speaking on behalf of the National 
Institutes. 

In order that we have a record which is clear, do you at the Na- 
tional Institutes of Health endorse the statement that was made by 
Dr. Broder? 

Mr. Varmus. Yes, I do. 

Mr. DiNGELL. You do. So that essentially becomes the statement 
of both NCI and the National Institutes of Health and the policies 
therein, then, as enunciated by Dr. Broder, the policies of NIH. Is 
that right? 

Mr. Varmus. In principle, of course, the specifics 

Mr. DiNGELL. I was comfortable till you said it "in principal", at 
which point my discomfort level rose. 

Mr. Varmus. I would just make clear that the specific remedi- 
ation procedures that Dr. Broder was specifying apply to his insti- 
tute. Other institutes have very similar measures. They may not be 
identical. I'd want to check them. 

Mr. DiNGELL. I think you are compelling us to inquire, then, as 
to the similarities and to get some appreciation. Would you submit 
to us the remediation procedures at the different institutes so that 
we can see whether, in fact, they meet your test or Dr. Broder's 
test? 

Mr. Varmus. We can do so. 

Mr. DiNGELL. It may be that there are some who are, let's say, 
not being as vigorous as you and I would like them to be and I 
think we'd like to have a little look at that. 

Mr. Varmus. May I say, Mr. Chairman, that Wendy Baldwin, the 
Deputy Director for Extramural Activities, who is sitting behind 
me, was asked by me a few weeks ago to look at the other insti- 
tutes' activities. I've been very pleased with what she's found and 
I think you will agree that they meet the high standards you would 
like to see in place. 

Mr. DiNGELL. Thank you. Doctor. Dr. Broder, the representative 
of the Breast Cancer Coalition complained, and I think properly so, 
about the lack of information regarding the B-14 treatment trial. 
Just last Friday, FDA and ICI/Zeneca, the manufacturer of 
tamoxifen, issued a new warning and label based, in part, on the 
findings of at least four cancer deaths associated with the adminis- 
tration of tamoxifen in that trial. 

The subcommittee has recently obtained from Dr. Fisher a chro- 
nology of events related to the deaths of these women and submit- 
ted it to NCI for review. What conclusions did NCI draw from your 
review and the analysis of the timeliness of the disclosures of the 
deaths of those patients? 



56 

Mr. Broder. Sir, members of your committee were kind enough 
to give me some information which, in part, we did not have. I was 
able to review them last night. 

It is my professional judgment that we should have received in- 
formation on certain facts and, particularly, if it would be possible, 
to have information on endometrial cancer early in 1992, possibly 
earlier, and that information was not provided to us until substan- 
tially later than that. 

This speaks to the issue of us wanting to hold the IND for such 
studies, which we did not have. But there was no information pro- 
vided and I believe that early in 1992, certain information about 
an endometrigd cancer death could have been provided to us. 

Mr. DiNGELL. So it would be fair to say that one of the deaths 
should have been understood and reported in early 1992. Is that 
fair? 

Mr. Broder. That's correct. I'm trying to be fair and physicians 
may have their own interpretation. This particular patient was ini- 
tially signed out as a pulmonary embolism, or at least that's the 
facts as I know them. Since the diagnosis had pulmonary embolism 
in it as the cause of death, I believe there was some legitimate 
basis for not making a decision. But I personally believe 1992 
would be fair. 

Mr. DiNGELL. Without objection, we'll put the chronology of that 
in the record. 

[The information follows:] 



57 



i-> ;: z 



ft s 



0> 



? 2 O 

£ — . uj 

= Z O = 

*^ o >- es 

S UJ ^ >- 



01 

fM 






(SI 












fM 



0. 0\ 



(M 

» 

(M 

I 












(M 



K- e Kj 

Wfl (1 Z 

s _ >^ 

- l^ M 



ss 


m — IV 

a> CA B> 


ea o lo (N — 
sa 0> 91 » ON 


? 


a cji m — 

G9 GS » CI 


ot u> 
1 1 


— a 


1 — e a 
1 a — to 


N « ~ e e 
1 (M — o in — 


. a 1 
* »- 1 


1 pg B ' a 
1 pg Kl a — 


1 m a 


s: 


e e B 


— O (M — — 

a — s e « 


a 


a a a a 


— pg 
a a 






s§ 



2= - 

K i u s 

>- (J O taJ 

M w a e 
= -I Z a 



pg — K> PM a pg Ki 
at ON 0« ot 0> e> *> 



0< PM — a 



r- <r m o> 



I I I I 



►n a Bi -• 
Q^ a* o^ *^ ' 



tz 



«i lo m pg e« vD vo 



pipgoS; — pgfoin 

_ 1 I I Wl I 1 I I 

R-.SSSpgSrvSa-S-BgNBBB 
-■-— to — efiS — (M 



et 10 tn 



^ a — — k Z I 



N r- 



1/1 



>£> 5 = B - 



pg a pg — 
^ a> » o> 



pg trt 

. . - . "» ^. • 
pg p«( pg <o (o It 



pg — 
01 01 



— a (M — pg pg o 
B> et ct 0t 0< 0> 01 



mwwmoiieot — 0ipJ>'ipgsSP''|g 



Z w 



s 


CM — a — miaoipg — apgMioin«« 
ea0i9oichaa0i0>0i0>9at0\aB 


a 
pg 


iaetpgpgtar'«ae><£csa — BtiaaiPi 
m — e — pg-'saae — loaaaip)-- 


01 


TJ.4---p:-ip!i-«^2!C!2a2a 



■t a> 01 a pg P>J d 
a a 01 9 0t ot 



m 01 ID v« 

— M — a 



o a 
Z Z 




58 

Mr. DiNGELL. Now, is it fair to say that the NCI and the manu- 
facturers should be notified in 1993 about a second death? 

Mr. Broder. That's fair. 

Mr. DiNGELL. The notification of the third death should have oc- 
curred by at least January 1993, isn't that so? 

Mr. Broder. I believe that that's fair. 

Mr. DiNGELL. So you have three and then the fourth death, there 
should have been a notification by at least August of 1993, isn't 
that right? 

Mr. Broder. Sir, I believe that's correct, but if that specific 

Mr. DiNGELL. I'm not going to hold you to something. You con- 
sider yourself free to respond to that question later, as the need 
might occur. 

Mr. Broder. Yes. Recognizing that there might be a minor devi- 
ation from that specific date. But I do take your point and I believe 
it's substantively correct. 

Mr. DiNGELL. When were you notified. Doctor? 

Mr. Broder. Sir, we received our notification through an NSABP 
meeting that was held in October — toward the end of October of 
1993. 

Mr. DiNGELL. Substantially later than the time that you should 
have been notified. 

Mr. Broder. You are correct. 

Mr. DiNGELL. Doctor, is it your testimony that the American pub- 
lic should be informed about the deaths associated with tamoxifen, 
as well as increased cancers associated with tamoxifen in 1992, not 
1994? 

Mr. Broder. I believe that an appropriate notification should 
have started early in 1992. 

Mr. DiNGELL. Here we have a bunch of women who have partici- 
pated in a test. They were advised of certain matters in connection 
with this to achieve their informed consent. The testimony so far 
indicates that they may not have been properly and adequately ad- 
vised to give their full and adequate consent based on a fair expo- 
sition of the facts and the risks and the data available to the exam- 
iners. 

Is that a fair statement? 

Mr. Broder. I believe it is fair. But if you will permit, the situa- 
tion has a complexity in that the very first informed consent did 
disclose to patients the possibility of lethal outcomes. They were re- 
lated to embolic, thrombophlebitic and cardiovascular. So the pa- 
tients were on notice that death was a possibility. 

The patients were also on notice that endometrial cancer was a 
possibility. But you are quite right and I believe your point is well 
taken that the issue of endometrial cancer, which should have been 
brought up to current state, was not provided. 

Mr. DiNGELL. The risk with regard to uterine cancer was either 
not stated or was significantly understated. 

Mr. Broder. The risk of death from uterine cancer most as- 
suredly was not properly and adequately provided. 

Mr. DiNGELL. As a matter of fact, was there any information at 
all with regard to possible fatalities resulting from this? 

Mr. Broder. Death to certain 

Mr. DiNGELL. From uterine cancer. 



59 

Mr. Broder. From uterine cancer, no. But death due to other 
causes was — the patients were put on notice that there was a pos- 
sibility of a lethal outcome from exposure to tamoxifen. 

Mr. DiNGELL. Is there any requirement now in the protocols or 
the rules and regulations of either NCI or NIH that would require 
women to be properly notified in connection with the risks? 

Mr. Broder. What we are doing now is that we are reexamining 
and will, as a condition of a grant, require the same kind of notifi- 
cation that the grantee would owe to the IND holder. We want to 
be in the information loop immediately, not as an afterthought. We 
are prepared to not fund proposals where that is not possible. 

In a community-based study, which is far flung and, by defini- 
tion, has a number of sites at multiple areas and also has central 
pathology services and other things that need to be done, one has 
to accept a certain amount of delay. Also, there is a need to make 
sure that if an alarm is sounded, that it is based on very valid in- 
formation so that the meaning of the alarm would be preserved. 

With all of that, I think that our processes need to be improved 
and we are taking steps to make sure that the grantee understands 
that. We will also strongly discourage either the private companies 
or the individual grantees from holding the IND, which, in effect, 
was the case here for the tamoxifen studies. 

Mr. DiNGELL. Doesn't it — I'm sorry. Dr. Varmus, go ahead. 

Mr. Varmus. It is, of course, NIH policy that all of our consent 
forms reveal all of the relevant information that would constitute 
risk to patients enrolled in these studies. We ask our Office of Pro- 
tection from Research Risks and the Institutional Review Boards, 
which include members of the public, often consumers who are in- 
volved in these studies, to be sure that the consent forms do ade- 
quately describe all of the known risks. 

Mr, DiNGELL. Have all of the women who were put at risk by the 
failure to adequately disclose the risk levels with regard to uterine 
cancer or other perils been informed of the new level of risk or the 
level of risk as we now know it? 

Mr. Broder. I'd like to ask Dr. Friedman to comment on where 
we are with that process. 

Mr. DiNGELL. They either had been advised or they had not been 
advised. We can get, I think, a fairly simple yes or no answer. 

Mr. Broder. In that case, the answer is no. 

Mr. DiNGELL. The answer is no. 

Mr. Broder. The process in large, far-flung, community-based 
studies requires several steps. We could theoretically provide notice 
of some type. A local institutional review board not responding to 
the government must approve and endorse any changes in in- 
formed consents. 

We do try to make the information as publicly available as we 
can, but I cannot assure you that every single woman has signed 
a new informed consent. 

Mr. DiNGELL. I'm trying to understand. We have many questions 
here, one of which is the simple moral question. Shouldn't these 
women be informed of the fact that they have been put at addi- 
tional risk which was not disclosed to them at the time they agreed 
to participate in the study? 



60 

Mr. Broder. Yes. And that process is going on now and will be 
completed. If your question is will every single woman be notified, 
the answer is yes. 

Mr. DiNGELL. When will that occur? Because they are now all at 
additional risk with regard to uterine cancer, some also with re- 
gard to blood pressure and possibly other matters. When will they 
receive notice? 

Mr. Varmus. Mr. Chairman, all of the 309 sites of the tamoxifen 
prevention study have been informed of the recommendations to 
change the consent forms to be in accord with the newly available 
evidence. Now, of course, it is the responsibility of those who are 
in charge of the study at each of those sites to respond and to in- 
form each of the patients. 

The NCI itself, of course, can't inform the patients directly. We're 
informing them through the several hundred sites that are partici- 
pating in this multi-institutional study. 

Mr. DiNGELL. Have you told me that all of the women have been 
informed or just that the sites have been informed? 

Mr. Varmus. All of the sites have been informed and we hope, 
we trust, and we will enforce the 

Mr. DiNGELL. One of the great charities or one of the three great 
virtues. But this committee has always, in dealing with matters of 
this kind, sought greater res^^onse from the government than sim- 
ply trust in the great virtues. 

I'm curious. How are we to say that the sites have informed the 
women? We cannot say so, can we? 

Mr. Broder. Mr. Dingell, you are quite right to raise this issue. 
The trial is currently suspended. We have taken our process of how 
one informs women of new and evolving side effects to the Office 
of Protection from Research Risks, which advises us in that proc- 
ess. 

Perhaps you might also wish to take into account the fact that 
in any study there will be an ongoing process of new risks, some 
of them acute, some of them longstanding, and some of them chron- 
ic. There's always going to be an updating process. So this is not 
the only time. We're going to have to continuously modify our in- 
formed consent. 

The Office of Protection from Research Risks advised us that 
upon their evaluation, this process did not require an immediate 
call-back notification, but could be done in the normal cycle in 
which women would come back to clinic. We, however, have chosen 
not to take that option and we will be calling women back. 

In addition, we will have a process to ensure that sites have noti- 
fied all women before that site can re-accrue women in the process 
when the current suspension is over. 

Mr. DiNGELL. I hear a goodly number of things, but we have es- 
tablished clearly that we cannot say that the women have been no- 
tified. 

Mr. Broder. I cannot tell you that every single woman has been 
notified. 

Mr. DiNGELL. We've also come to the conclusion, and I think you 
concur in this. Dr. Varmus, that there is a moral requirement that 
the women who participated in this study should be given correct 
information about new risks which have come to light since the 



61 

time that the study was initiated and since they were advised of 
the level of risk. Is that not so? 

Mr. Varmus. Yes, I agree with that. 

Mr. DiNGELL. And we can establish that we do not know, but we 
hope that the women have been informed by the sites. Now, we can 
observe fairly that the women involved in the test are probably still 
taking the drug, can we not? 

Mr. Vaemus. Mr. Chairman, we will be in touch with every one 
of those sites to ensure that every woman who is enrolled at those 
sites has received the information. 

Mr. DiNGELL. That has not yet occurred. 

Mr. Varmus. That's correct. 

Mr. DiNGELL. We also have a modest little problem in the fact 
that the women should be informed so that they can know what 
the new level of risk might be. We also had the additional problem 
that there is no requirement in the rules or regulations of NIH that 
the women do receive notification of changes in the level of risk to 
them from this particular test program of which they have become 
a part. Is that not so? 

In other words, there is no requirement in the program that the 
women be warned of changes in risk as they occur to them or in 
incorrect representations to them of the level of risks which were 
made early on to induce their participating the program. 

Dr. Chabner, do you want to comment on that? 

Mr. Chabner. Yes. There is an absolute responsibility to notify 
participants. 

Mr. DiNGELL. You agree there's a responsibility, but what is 
there in the regulations? 

Mr. Chabner. Yes. The IND holder must notify participants in 
a trial. 

Mr. DiNGELL. You can, but this has not been done and it is not 
in the regulations at NIH to require the women to be informed, is 
it? Yes or no? 

Mr. Chabner. I'm not sure. 

Mr. DiNGELL. Well, that's a good answer. It tells me that there 
is no requirement at NIH. 

Mr. Chabner. I know it's a requirement for the IND holder. 

Mr. DiNGELL. Maybe Dr. Friedman or Dr. Broder or Dr. Varmus 
can tell. 

Mr. Broder. Mr. Dingell, I'll shortcircuit the issue. I will just 
speak for the NCI. There is a requirement. 

Mr. DiNGELL. There is a requirement? 

Mr. Broder. I'm saying there is a requirement. 

Mr. Dingell. Has that requirement, then, been honored here? 

Mr. Broder. It has not been honored from the point of as we 
speak today. 

Mr. DiNGELL. But as we speak today, women are taking 
tamoxifen. And as we speak today, women are not being informed 
of the level of risk that they confront. As we speak today, women 
are not able to go to their own doctor and say ,"Doctor, should I 
continue in this research program with a potentially increased level 
of risk." Nor are they being informed that they can have their own 
doctors inquire and to monitor with greater care into their level of 
safety. 



62 

Mr. Broder. If I could say, the methods that we — we cannot, 
under current operating rules, command an institution to adopt our 
informed consents. But we will reexplore that issue even. What we 
are trying to do is — the process involves sending out changes. One 
of the points I would like to stress is that there will be many 
changes. There are new things that one learns in any study. 

So we have a process. This will not be the last time we will 
change this informed consent or others. What we are trying to do 
in the interim is notify women in a number of ways, notify women 
through the cancer information service, which women can call, no- 
tify them through our electronic data boards. A letter from Zeneca 
is in our PDQ and cancer facts system. 

In other words, we are doing what we can to get the word out. 
We will have rules in place so that accrual cannot start and insti- 
tutions will need to document for us that all women have been in- 
formed. But you are right that the process has a certain delay ele- 
ment in it. 

Mr. DiNGELL. But here you have the good Dr. Fisher, the Univer- 
sity of Pittsburgh, all the grant recipients, all the folks who are 
participating in the study, and you have imposed no requirements 
on them, as the record stands at this particular time, requiring 
them to report to the participants in the study of new information 
relative to risk, nor of increased risk which has come to your atten- 
tion. 

You're doing all this wonderful work and I appreciate that, but 
you're not imposing any requirement on the dear University of 
Pittsburgh or the good Dr. Fisher to get this information to people 
who may potentially be put at risk by the changes in the cir- 
cumstances, by the new information, and by the failure to ade- 
quately warn them back in 1992. 

Is that a fair statement or not? 

Mr. Chabner. Congressman, the University of Pittsburgh, Dr. 
Fisher, holds the IND in this trial. They have a legal obligation to 
notify participants in their trial. That is an obligation. 

Mr. DiNGELL. Have they done that? 

Mr. Chabner. They've been instructed to do it. Whether they've 
done it or not 

Mr. DiNGELL. They seem to be somewhat deficient in following 
up on instructions, according to what we've gotten today. 

Mr. Broder. We take your point. 

Mr. DiNGELL. Remember, the IND, that's the investigation on the 
new drug, is that right? That is not an NIH requirement. 

Mr. Broder. That's correct. 

Mr. DiNGELL. That is a Food and Drug requirement, right? 

Mr. Broder. Yes. But as a requirement to get the grant, they 
must have the IND. 

Mr. DiNGELL. Pardon? 

Mr. Broder. It is our requirement, in order to do the study, they 
must have an valid IND. So if their IND is revoked or improperly 
being performed, then the study must stop. 

Mr. DiNGELL. It probably is too much to say they've been snap- 
ping their finger under your nose, but certainly their behavior has 
been less than the requirements that you would impose on them 



63 

in terms of good moral behavior or in terms of seeing to it that the 
experiment is conducted in a sound scientific fashion. 

Mr. Broder. They don't respond to constructive criticism. 

Mr. DiNGELL. Well, we're going to help. We think that now we 
maybe ought to have the University of Pittsburgh before us to dis- 
cuss these things. Maybe we might have you and Dr. Varmus back, 
and maybe Dr. Chabner and Dr. Friedman and maybe even Dr. 
Bivens, because we sense that matters are not being handled in a 
way which makes this information fully available to the women 
who have been involved in the test. 

The women are already in the treatment trial. Many have taken 
the drug for as long as 10 years. 

Mr. Broder. For therapy of breast cancer, that's correct. They 
take it for multiple years. 

Mr. DiNGELL. And some are taking it as a preventative. Now, 
there appears to be strong evidence that, in point of fact, it's not 
a good preventative and there also appears to be some significant 
evidence that it imposes additional risk stemming from potential 
other perils; i.e., uterine cancer and blood pressure phenomena. 

Mr. Broder. May I respond to that? 

Mr. DiNGELL. But neither you nor Dr. Varmus are able to tell us 
that this information has been made available to the women who 
are involved in the study. We are trusting the Food and Drug Ad- 
ministration, which, not infrequently, we have before this commit- 
tee to discuss some of their shortcomings. 

Mr. Broder. Mr. Dingell, I take your point. Perhaps, with your 
permission, I'd like to ask — well, Dr. Ford has not been sworn in. 
I will certainly provide to you for the record the steps that we have 
taken. 

But if you will permit me a very brief interval. You made a state- 
ment which I would argue is still open to some discussion. We un- 
derstand, respect and fear breast cancer. It is an invasive disease 
and a woman who develops breast cancer has, in spite of all the 
best things that we have to offer and we have had improvements, 
but on average, a woman who develops breast cancer has approxi- 
mately a 50 percent chance of dying of her disease. 

I come to you to this table with an appreciation of the formidable 
quality that breast cancer has. We do not have any immediate pre- 
ventions to offer women who are at high risk. Some women, under 
50 and over 50, have risk factors which we can identify that impose 
an enormous burden of the possibility of breast cancer for them 
and they don't have any options. 

What we are saying by the tamoxifen study is that certain 
women who have high risk of breast cancer can possibly — we will 
possibly learn whether tamoxifen could be a preventative for those 
women. We have certain things, like hormonal replacement ther- 
apy, which is available in this country in any drug store by a pre- 
scription, which causes certain benefits to women who are post- 
menopausal, but which is associated with a comparable 
endometrial cancer risk, and it's never been the subject of a clinical 
trial. 

All we are saying is that we understand the problems, but 
tamoxifen has been one of the few things in the scientific literature 



64 

which has a proven and multiply replicated reduction of 
contralateral breast cancer in women who already have had 

Mr. DiNGELL. Doctor, we do not talk about it as a useful treat- 
ment device for breast cancer. 

Mr. Broder. Even for prevention. 

Mr. DiNGELL. I do not quarrel about that. 

Mr. Broder. What I'm talking about is it is — what we have 
learned is that women who have breast cancer have a separate and 
independent risk not only of recurrence, which is one problem and 
is the most often thing we're talking about, but they have an inde- 
pendent risk of developing a new breast cancer in the opposite 
breast. 

Tamoxifen has been shown to reduce that in several studies by 
a substantial amount. Therefore, it also possibly may have certain 
effects for myocardial infarctions and so on. Women who already 
have breast cancer who have received tamoxifen have fewer heart 
attack rates than the control group. 

We are not saying that tamoxifen should be used. We ask for it 
not to be used for prevention. What we are saying is for women 
who are at risk, we ask for the opportunity to do a clinical trial 
so that we can inform you and the public as to whether this is an 
option. That's all we are saying. 

Mr. DiNGELL. Let us lay this matter to rest. I am not quarreling 
with the fact that you're performing the test. I am trying to ad- 
dress the reporting, the disclosure and the information which is 
given to the women who are involved in the particular test. 

We've talked here about reports with regard to treatment and 
with regard to prevention. But I'm trying to find out the reports 
that you have received in each instance and the reports that the 
women who are involved in these programs are receiving. I sense 
that they may be getting certain reports in connection with the 
treatment program, but they may not be getting it in connection 
with the prevention tests which are going on. 

Now, am I fair in that inference or not? 

Mr. Broder. Yes, you are. You are fair and within 24 hours we 
will give you a report of the status of what our information flow 
has been. 

Mr. DiNGELL. That would be very helpful. My time has expired 
and I want to recognize my good friend and I'm going to apologize 
to him right now, but there's a question that we have not gotten 
an answer on. We do not now know whether the women have re- 
ceived the necessary notification of either the new risks, the new 
circumstances, or the failure to adequately inform them. 

We do not know yet whether or not you have in your rules and 
regulations and in the protocols the requirement that kind of infor- 
mation be made available. We do not know whether that kind of 
information has, in fact, been made available and whether there is 
a policy change needed at NIH to see to it that this notification 
goes forward to the people who have been put at additional risk be- 
cause of the kind of situation we find in the tamoxifen tests. 

Can you tell us whether or not you have requirements in the pro- 
tocol, either at NIH generally or at NCI, which requires that this 
new information be made available to the persons who were in- 
volved as it comes available? 



65 

The problem we have here is women are at risk. Women are sub- 
ject to potentially higher levels of risk of cancer, uterine cancer, in 
connection with it. We also want to know whether they are able, 
then, to monitor their own affairs more fully and carefully because 
they have the risk that they can take to their own doctors or that 
can be made available to the researchers who might be working 
with them at the different sites. 

Can you address those questions, please? 

Mr. Varmus. Mr. Chairman, let me see if I can help with this. 
Under the Office of Protection from Research Risks, we have rules 
that dictate that information that concerns adverse outcomes or ad- 
verse effects of instruments used in clinical trials or prevention 
trials be communicated to the institutional review boards at the 
sites at which these studies are being conducted. 

Mr. DiNGELL. That's to the people at the site. 

Mr. Varmus. Let me finish, please. At those sites are found the 
names and addresses of study participants. We do not have at NIH 
the names and addresses of participants. 

Mr. DiNGELL. I'm not critical of NIH with regard to your inability 
to notify these people. I am trjdng to find out whether you have a 
situation which does require the necessary notification. 

Mr. Varmus. It is a regulation that all patients in the study be 
informed through the institutional review boards. You're raising a 
legitimate question of whether we should have tougher rules to fol- 
low-up on the notification process. You can rest assured that we 
will be looking into that very vigorously. 

Mr. Broder. I'd like to second that, because I want to assure you 
that we take your concerns and criticisms very seriously and to 
heart. I believe they are valid. We will look to see whether there 
is a structural problem that we can repair. We will also explore, 
although I cannot promise you today, whether at least the NCI has 
some right of direct access to patients. I believe that we will prob- 
ably be told that we have limited access. 

Mr. DiNGELL. I think it would be useful if you had direct access, 
but I think it's even more useful if you could simply require that 
the patients do be informed. 

Mr. Broder. That is done. 

Mr. Varmus. We do require that. 

Mr. Broder. That is required. 

Mr. Varmus. But you're asking whether I know that all the pa- 
tients have been informed and I cannot say that I know that. 

Mr. Broder. Also, we have limited rights of asking a grantee to 
provide names and addresses, but we will certainly try to review 
that, as well. 

Mr. DiNGELL. The deaths should also have been reported in 1992 
before any of the patients entered the prevention study, should 
they not? 

Mr. Broder. That's correct. 

Mr. DiNGELL. That was not done. 

Mr. Broder. That was not done. As far as I can tell, it was not 
done even to the FDA. 

Mr. DiNGELL. Can you tell us why? 

Mr. Broder. I would have to speak for Dr. Fisher and I would 
prefer not to do that. 



66 

Mr. DiNGELL. Well, it might be a little bit difficult. As the record 
shows, we have invited Dr. Fisher to appear here and he has indi- 
cated that his health prevents him from being with us. We perhaps 
will be affording him another opportunity, but as I have indicated, 
we are going to be asking the university to come forward. 

Mr. Broder. I hope his health improves. 

Mr. DiNGELL. Pardon? 

Mr. Broder. I hope his health improves. 

Mr. DiNGELL. I do, too. At least enough that he can appear here 
before us. Well, the Chair has used more time than I am entitled 
to. The Chair recognizes the gentleman from Colorado, Mr. Schae- 
fer. 

Mr. Schaefer. Dr. Broder, the NCI issued a statement on March 
14 of 1994 that I would want to read for the record. I'm sure you're 
familiar with it. Its headline is "National Cancer Institute's State- 
ment on Breast Cancer Treatment Studies." 

The original conclusion reads as follows. "Following evidence that 
fraudulent data has been submitted as part of large breast cancer 
treatment group trials, the scientists overseeing the trials and staff 
of the National Cancer Institute reanalyzed the data. The re-ana- 
lyzation was done without any of the data supplied by the inves- 
tigator accused of fraud and that reanalysis using data exclusively 
from other institutions participating in the group reaffirmed origi- 
nal results." 

Was this a true statement when it was issued? 

Mr. Broder. I believe it was substantively correct. There are 
many issues about what the word "reanalysis" meant. But there 
was in hand at that point a report from the NSABP which had cer- 
tain issues of discussion, but which substantively confirmed the 
end points and did remove the fraudulent data site. 

Mr. Schaefer. Did the NCI analyze this data? 

Mr. Broder. The NCI did not analyze the data. The people 
overseeing the study that I believe you read were the people from 
the NSABP, the principal investigator. The NCI analyzed the data 
from the standpoint that our statistician, a government statistician 
who critically reviewed the report, made suggestions that were 
transmitted back to the NSABP, but, in fact, did feel that the sub- 
stantive issues were valid, that the substantive end points were 
valid. 

Mr. Schaefer. But NCI was forced to retract the statement after 
it was released. Is that correct? 

Mr. Broder. It became a definitional term about the word "ana- 
lyze." I believe that people of goodwill can use that term in mul- 
tiple ways. We understand that term had a special meaning and 
when individuals meant the term "reanalyze", they wanted us to be 
the individuals to reanalyze it. 

I'm sorry for the confusion. It will not happen again. 

Mr. Schaefer. What about the ordinary people who were in- 
volved in reading it? Did they understand this stuff? 

Mr. Broder. The word "analysis" is open to multiple interpreta- 
tions. One can analyze a scientific experiment or analyze some- 
body's argument or analyze a problem without necessarily doing 
the firsthand work oneself One looks at the work product and then 
analyzes it. That was the use of that term. 



67 

We're sorry if there is confusion on that point, but we do have 
our own analyses which have been provided to the committee. So 
any doubt or ambiguity on this point has now been laid to rest. 

Mr. SCHAEFER. Let me ask you this question. How many people 
received this erroneous statement? 

Mr. Broder. Sir, with respect, this is open to different interpre- 
tations. If you believe the statement was erroneous, I will accept 
your point. But I believe the term "analyze" is open to multiple in- 
terpretations. 

Mr. SCHAEFER. But you still had to retract it and then reanalyze 
it, didn't you? 

Mr. Broder. But we have the reanalysis, sir. It's been provided. 
Our own analysis of the relevant studies has been provided to the 
committee. It is on Internet, it is on our electronic databases. It has 
been sent to members of the National Cancer Advisory Board. It's 
been sent to Kay Dickerson, it's been sent to Fran Visco and so on. 

These are our own analyses, not the NSABP. We have distrib- 
uted this and we are sorry that there mav have 

Mr. Schaefer. This is an updated analysis now. 

Mr. Broder. No. It is our work. We asked for and demanded the 
original computer data files, something which is done extremely 
rarely. It will be more common in the future. But we asked for and 
were provided the original data files. We are also doing chart au- 
dits, which is different from the computer data files. But those doc- 
uments have been provided. I can provide them to you now, if you 
will permit me to put them into the record. I have them with me. 

Mr. DiNGELL. Without objection, so ordered. 

[The documents referred to are retained in subcommittee files.] 

Mr. Broder. With the Chair's permission, I will provide you with 
the documents as they are being distributed. 

Mr. Schaefer. I really want to get down to this point. The first 
statement was issued. It was retracted, or it was redone or 
reanalyzed. What was it? There was a difference, right? So, there- 
fore, people read the first statement and how did they 

Mr. Broder. And they're true and they say different aspects of 
it. In the early portion of February, we received an NSABP report 
which analyzed, statistically analyzed the end points of the various 
studies affected by the fraudulent data site, with and without the 
data site from the fraudulent site in the report. 

With the Chair's and the committee's indulgence, I have to intro- 
duce a term. They analyzed it by intent to treat, which is a valid 
way, but did create some confusion. I can define that term now or 
I can define it to you for the record. 

But it does provide a valid way of analyzing material. That was 
reviewed by our in-house staff and it was reviewed by Dr. Simon, 
who raised certain criticisms, constructive criticisms that were pro- 
vided. But the conclusion was that the Poisson data removal did 
not change any of the major end points. In fact, the studies basi- 
cally, in most of the major things that we discussed, were the 
same. 

We have, in addition to that, demanded the original computer 
data files and have run with our own agents a reanalysis independ- 
ently down here in Bethesda. Those reports are with me today and 
I will be glad to submit them to you for the record. They have al- 



68 

ready been circulated. They are, as we speak, being circulated. 
They are on Internet. They are available by our — anyone who has 
a computer service can download this through our PDQ system and 
so on. 

So what I'm trying to say to you is that this process has more 
closure than perhaps — it's called the EMMES analysis because this 
is a private, highly sophisticated, statistical firm that's a contractor 
that reports to us. They did everything from the computer data 
files. 

We also are auditing the various sites with our own government 
workers or contractors that report to us. Dr. Friedman is prepared 
to provide you with the audit summaries as to where we are today. 

So I apologize if the term "analyze" had some confusion, but to 
those of us in the scientific community, the word "analysis" or 
"analyze" has multiple interpretations. But all interpretations of 
that term have been resolved as we speak today. 

Mr. SCHAEFER. Let me ask you this question. Did the reanalysis 
review all the points in the same manner as the original analysis? 

Mr. Broder. It did it both as to the original analysis, but be- 
cause we had possession of the computer data file, we were able 
to ask and clarify questions that individuals have raised; for exam- 
ple, the ipsolateral breast cancer story we could address and we 
could do certain kinds of analyses that were not even in the origi- 
nal publications. 

So the answer is, yes, we did and we did more. So there is a 
value-added function in looking at the reports that we have and the 
reader will be able to get more information. 

We also have a copy of the New England Journal of Medicine 
submittal which we compelled Dr. Fisher to submit. I have pro- 
vided a copy of that New England Journal paper to the committee 
and would be prepared to distribute it to any members of the com- 
mittee who wish to have it. If the committee advises us to, we will 
make the New England Journal submittal public, as well. 

Mr. SCHAEFER. Who was responsible for releasing this statement 
in the first place? 

Mr. Broder. This was probably coordinated through our Office 
of Cancer Communication. I will accept responsibility for this and 
any other matters that the committee feels needs to be accounted. 

Mr. SCHAEFER. We did a re-analysis and I just want to say that 
for the common lay person out there, who may not understand sci- 
entific terms, that we would do well in the future to make sure 
that whatever is conveyed to these individuals, is not going way 
over their heads. They'll know exactly what was talked about. 

Mr. Broder. Sir, these will be the documents that will be sub- 
mitted so that they can be identified. These are the documents that 
we have done. I have with me here a copy of Dr. Fisher's New Eng- 
land Journal of Medicine paper. 

What I would like to advise the committee is that we have addi- 
tional analyses in our preparations, looking at different data points 
and different ways of looking at the data, that actually are not 
even considered in this paper. Dr. Fisher's paper. But the bottom 
line is that we have looked at all of the end points that we know 
how to look at and including some that he has not published. 



69 

Mr. SCHAEFER. Would you then say that the three ladies that we 
had here prior to this panel, that they would take those and could 
understand them? 

Mr. Broder. I believe so. I believe the panel was composed of ex- 
tremely intelligent women who have asked good questions and who 
understand the facts and are among the most tragic aspects of this 
issue. Basically, we need to understand at the National Cancer In- 
stitute that we report to the first panel and not to Dr. Fisher. 

Mr. ScHAEFER. Mr. Chairman, I'm over my time here. I have a 
couple other questions. 

Mr. DiNGELL. The Chair will recognize you further. 

Mr. SCHAEFER. Dr. Friedman, you were responsible for 
overseeing and monitaring the reports that came in from the Uni- 
versity of Pittsburgh, is that correct? 

Mr. Friedman. Yes, sir. 

Mr. SCHAEFER. How many people were assigned to this specific 
oversight? 

Mr. Friedman. There were two people at that time in the audit- 
ing section. Neither one was devoted specifically to the University 
of Pittsburgh. Both had responsibility for auditing and quality as- 
surance oversight. 

Mr. SCHAEFER. Is this an adequate number to correctly do it? 

Mr. Friedman. I think the problem was not so much the num- 
bers as the procedures and the ability that we exercised in having 
what Dr. Broder has described as constructive criticism actually 
translated into meaningful action. 

Mr. Schaefer. How can we assure that reports now coming from 
the University of Pittsburgh are being adequately monitored? Are 
you investigating whether you're going to use more people or not? 

Mr. Friedman. Sir, there are a number of answers to that ques- 
tion and I will be happy to give you the various aspects of it and 
then ask Dr. Broder or Dr. Chabner to supplement anything that 
I have left out. 

First of all, the entire way in which business is conducted, re- 
search business and quality assurance business at the University 
of Pittsburgh and the National Surgical Adjunct Program, has been 
changed. The laxness, the imprecision which characterized some of 
their administrative activities in the past no longer is allowed to 
exist today. 

So that they must provide to us and have provided to us a com- 
plete auditing schedule. They understand what their requirements 
are for reporting things to us in a timely way. By that, I mean they 
must report to us within 24 hours after an audit is conducted. We 
must have within 6 weeks a written report of that audit. 

Mr. Schaefer. When was all this changed? 

Mr. Friedman. Sir, these were new — when the University of 
Pittsburgh and this group were put on probation at the very end 
of March, a whole series of instructions were provided to them and 
those are the features that I'm speaking about, sir. 

Mr. Schaefer. So we had to have an issue like this come out be- 
fore you were moving to do new things on adequately monitoring 
this information. 

Mr. Friedman. Sir, the numerous times that we contacted 

Mr. Schaefer. We're dealing with a lot of women's lives here. 



70 

Mr. Friedman. I'm sorry, sir. I didn't hear you. 

Mr. SCHAEFER. We're dealing with a lot of women's lives in this 
country. 

Mr. Friedman. The subject is enormously important. The quality 
of these data, the trust with which the public holds these data are 
terribly important. Our ability to compel Dr. Fisher, our ability to 
institute changes that we had recommended did not occur until 
very recently. 

Mr. Schaefer. So you had recommended these changes prior to 
these statements and the studies being issued? 

Mr. Friedman. Yes, sir, we had. 

Mr. Schaefer. And it was stonewalled. In other words, nothing 
was done. 

Mr. Broder. I believe the correct answer is that Dr. Fisher, 
maybe not directly, but, in effect, told us by implication that he's 
been doing clinical trials a long time, perhaps before we were out 
of school. He knows what to do. He's been doing things since the 
late 1950's and he knows how to get the job done. 

No one will ever tell us that again. We will do what we have to 
do. Our staff understands the mission. They understand what we 
need to do. I believe that Dr. Fisher has made a number of accom- 
plishments and I believe his accomplishments have benefited 
women to an extremely high degree. That's not the issue here 
today. 

The issue here is that no one is above our rules and we will see 
to it. 

Mr. Schaefer. Well, I think that's something that the chairman 
and I are very pleased to hear. If I might ask one more question, 
Mr. Chairman. 

Mr. DiNGELL. The gentleman continues to be recognized. 

Mr. Schaefer. Dr. Broder, did any officials at NCI notify the edi- 
tor of the New England Journal of Medicine that the St. Luc data 
was in doubt? 

Mr. Broder. No, they did not, until very recently. 

Mr. Schaefer. Why weren't scientific journals notified? 

Mr. Broder. Because it was the mistaken belief that the primary 
obligation should be with the person who authored the words. I 
still hold to that principle. The person who authors a paper has the 
primary duty to retract and correct that paper, irrespective of what 
a government agency does. 

So even though we are changing and I will provide you with a 
different point of view, I still think that responsibility should not 
be lost sight of The primary duty for correcting a paper lies with 
the primary investigator who wrote the original paper. 

The belief, I believe, was that Dr. Fisher was extremely pre- 
eminent, extremely experienced, knew what he was doing. In addi- 
tion, there was some inhibition and self-consciousness about order- 
ing him to do anything. 

The other issue is that it would have been best for all parties had 
there been a simultaneous process in which Dr. Fisher and our 
group cooperated in revealing all of the facts as promptly as pos- 
sible. It becomes difficult and awkward for the NCI to act alone 
without the investigator cooperating with us. 



71 

However, that will not happen again. We will have in place a pol- 
icy in which these things occur promptly, even possibly and likely, 
in fact, in a clinical trial before the ORI investigation is over, and 
we have proven that. You have a press release today, I understand 
from Dr. Bivens. 

Mr. SCHAEFER. What is the press release? What does it deal 
with? 

Mr. Broder. Dr. Bivens? 

Mr. ScHAEFER. You're saying you're going to change the way 
you're doing 

Mr. Broder. No, no, no. It's a substantive press release. 

Mr. Bivens. I don't know if it has been released or not. 

Mr. Broder. The press release is being released today on a sec- 
ond case of fraud at 

Mr. Schaefer. A second case. 

Mr. Broder. Yes, at another hospital in Montreal. 

Mr. Schaefer. Dealing with breast cancer? 

Mr. Broder. That's correct. 

Mr. Schaefer. Or something else? 

Mr. Broder. Dealing with breast cancer. 

Mr. Schaefer. Well, just briefly tell me and Mr. Chairman here 
what this press release says. 

Mr. Broder. I believe the facts have been provided to the com- 
mittee. 

Mr. Schaefer. Do we have that? 

Mr. Dingell. There was a comment, I believe, in Dr. Broder's 
prepared testimony on this particular matter. 

Mr. Broder. I addressed this point in my opening statement. 

Mr. Varmus. This matter has been reported in the press pre- 
viously. There have been allegations of irregularities or there have 
been irregularities detected in recordkeeping in St. Mary's Hospital 
in Montreal. 

Mr. Dingell. Did that also involve Dr. Poisson? 

Mr. Varmus. No, it did not. 

Mr. Dingell. A different doctor. 

Mr. Varmus. It was a different case. It was a case that was most 
recently described in the press after the NCI visited the NSABP 
Center in Pittsburgh. The case was reported to the Office of Re- 
search Integrity and that office is investigating the nature of the 
irregularities and determining whether or not they constitute sci- 
entific misconduct. 

Mr. Broder. But Dr. Fisher was removed within days of the de- 
tection of the second issue. I apologize if my opening statement did 
not clarify all the facts, but I did try to address them in my open- 
ing statement and would be happy to provide more information. 

Mr. Dingell. I think more information on this matter would be 
useful, Doctor, and we'd appreciate that, for the record. The gen- 
tleman from Colorado continues to be recognized. 

Mr. Schaefer. I thank the Chair. Dr. Bivens, I noticed that Dr. 
Poisson was prohibited from serving on the Public Health Advisory 
Committee and was barred from receiving Federal funds or grants 
for a period of 8 years. Is this a maximum debarment? 

Mr. Bivens. It's very close to the maximum debarment that the 
Department imposes. 



72 

Mr. SCHAEFER. What is the maximum? 

Mr. BiVENS. The longest one I know of is 10 years. There may 
be longer ones, but the only one I'm familiar with, which is longer 
than 8 years, is another scientific misconduct case. The modal term 
for debarments across the Department is on the order of 3 years. 
So anything in excess of 3 years we have to make an especially 
strong case for to the debarring official. 

Mr. SCHAEFER. Is that regulation? 

Mr. Bivens. I don't know that the term of debarment is written 
into the regulation, no, sir. But it's a practice that the Assistant 
Secretary for Management and Budget applies, in which 3 years is 
the usual. If there are more serious offenses, it can be longer. I 
can't say much more than that because I don't want to speak for 
the Assistant Secretary for Management and Budget. 

Mr. SCHAEFER. I guess the question I have is that if somebody 
is guilty of fraud, why should they ever, ever be eligible to receive 
Federal grants? 

Mr. Bivens. That's a perfectly legitimate question. I think in 
some cases, they should never be eligible to receive grants. I guess 
I would be surprised of Dr. Poisson ever came in for an NIH appli- 
cation in the foreseeable future. But, nevertheless, I'm stuck with 
the practice of the Department and the standard terms of debar- 
ment. We have to argue quite strongly for debarments in excess of 
3 years. 

It is certainly arguable that 8 years is insufficient. 

Mr. SCHAEFER. I would certainly agree with that. I will yield to 
my friend here for a minute. Let me regroup and see what else I've 
got here. 

Mr. DiNGELL. I want to come back to the question. You had two 
studies here and we dealt with the question of notification. We 
have talked of the prevention study, but we have not talked to the 
treatment study. 

On the treatment study, has there been any notification either 
of NIH or of the participants in the study with regard to the possi- 
bility of increased risk of uterine cancer? 

Mr. Broder. With the Chair's permission, I'd like Dr. Chabner 
to address that. 

Mr. DiNGELL. Can you help us out with that, please, Doctor? 

Mr. Chabner. May I address that, sir? 

Mr. DiNGELL. Certainly. 

Mr. Chabner. Yes. On January 12, we issued new information 
to all of the treatment centers to inform them that informed con- 
sents must be changed to reflect new appreciations of what the 
risks of endometrial cancer were and the fact that we had just 
learned that deaths had been reported. 

Mr. DiNGELL. Was that done, do you know? 

Mr. Chabner. I do, sir, because 

Mr. DiNGELL. Was it done by the centers? 

Mr. Chabner. Yes, sir. I've gotten information back from our co- 
operative groups who participate in these — who sponsor and direct 
these studies that the amendments have been changed; that is, the 
informed consent documents have been changed for all the studies 
and that patients are being informed at this moment. 



73 

If I may point out, sir, some of the studies already had within 
them the risk of death and they had within them descriptions of 
endometrial cancer risks. So what we're doing is making sure all 
the studies come up to a certain standard. 

These studies are slightly different than the prevention trials, 
sir, in that this is the commercially-available use of this agent; that 
is, tamoxifen is commercially-available and many thousands of 
women take it. So we are trying to have our informed consent docu- 
ments measure up to the latest information that we get from the 
drug company, as well. 

Mr. DiNGELL. I applaud this. We've come back to the prevention 
study that we have not been discussing at this minute. We still 
confront the problem that we don't know what has happened with 
regard to the women who are participants in that study in terms 
of upgrading the notice that they had of potential risks. Is that 
right? 

Mr. Chabner. Sir, my understanding is that at all the sites, in- 
formation is provided for both the prevention trial and the treat- 
ment trial. Just as others have said, I cannot assure you today that 
every last woman has been informed. 

Mr. DiNGELL. I accept that and I don't want you to feel I'm inter- 
rupting you. I'm satisfied with your answer. But we then confront 
the little additional difficulty that we don't know yet what the re- 
quirements are with regard to warnings to participants in the stud- 
ies. 

Am I fair in that appreciation. Dr. Varmus or Dr. Broder? 

Mr. Varmus. We do require that they be informed. But you are 
asking whether we know that everybody has been informed and we 
do not know that. 

Mr. DiNGELL. But what's in the protocols with regard to inform- 
ing participants in these studies about changes in circumstances; 
in other words, additional risks that might become known as the 
process goes forward? Do you have some kind of a generic policy 
with regard to that or do you not? 

Mr. Broder. While they're looking for the documents, I will 
speak for the NCI. There is a requirement that institutions, as part 
of their human subjects assurance, a broader policy than the NCI, 
must have a process for informing patients about research risks 
and benefits, that they must have an institutional review board 
that must approve such changes, and that they will update and 
bring new information both to the institutional review board and 
to the patients as new facts become available. 

Mr. DiNGELL. So we would have a problem, then, if that kind of 
notice had not been given to the women who were participating in 
the studies. 

Mr. Broder. We have a problem whenever information flow is 
blocked. 

Mr. DiNGELL. Could you check and find out for us, please, gentle- 
men, whether or not the information on these matters has flowed 
through in compliance with the regulations at NCI? 

Mr. Broder. I will give you a status report within 24 hours. 

Mr. DiNGELL. It doesn't have to be 24 hours. Just at your earliest 
comfortable convenience. We're speaking here as friends, we want 
you to understand. 



74 

Mr. Broder. We appreciate that. 

Mr. DiNGELL. This to Dr. Varmus. Doctor, how seriously do you 
view the withholding of information about tamoxifen-related deaths 
by one of NCI's largest grantees and one of the largest grantees 
from the FDA and potential withholding of information from NCI, 
potential withholding of information from FDA, and from the 
American public? Is this a serious matter or not? 

Mr. Varmus. Of course, it's a serious matter, extremely serious. 

Mr. DiNGELL. Are you concerned? 

Mr. Varmus. Of course, I'm concerned. 

Mr. DiNGELL. Figure we maybe ought to take a look at the rules 
and regulations and see whether they're inadequate? 

Mr. Varmus. The regulations with respect to the reporting of in- 
formation to us, I think, is unequivocal. What we're trying to estab- 
lish here, because you've raised an interesting question, is the reg- 
ulations as they pertain to our assurance that any individuals en- 
gaged in any clinical research studies be informed through the con- 
sent form of any changes in our information. 

There is, it appears, still a degree of local autonomy with respect 
to how the consent forms are constituted. We are obligated to in- 
form the institutional review boards of any new information that 
is pertinent to the writing of the consent form. It is then in the ju- 
risdiction of that institutional review board to review the informa- 
tion and to make a decision about how they're going to change the 
form. 

The form must then be returned to us so that we can review it 
and advise them. But you've raised a question that is new to me, 
frankly, with respect to the degree to which we can impose our will 
upon the rewriting of a consent form in response to new informa- 
tion. There is, at the moment, some autonomy accorded to the local 
institutional review boards. 

Mr. DiNGELL. Let's get down to a document that has just fallen 
into the hands of the committee. Yesterday afternoon, Zeneca Phar- 
maceuticals produced a number of documents to the subcommittee 
regrading their knowledge of cancers and deaths associated with 
the B-14 trial. One such document is an internal Zeneca memoran- 
dum, dated July 7, 1993. I'm going to quote from parts of it here. 

It says as follows, and I'm quoting, "I then proceeded to tell Dr. 
Fisher that the increasing number of patients within the B-14 trial 
developing endometrial cancer while on Nolvadex did prompt us to 
look at this issue more closely. Upon careful review of the data, we 
felt that there was an increased risk to develop endometrial cancer 
with Nolvadex and that we will modify our label to reflect this. I 
also commented that it was impossible to precisely quantitate the 
relative risk, although, qualitatively, it did appear that there was 
a slight increase incidence." 

"Dr. Fisher agreed with this assessment and felt that we were 
acting responsible", and that's a direct quote, "by changing our 
label. I commented to him that at least in the United States we 
would have to make minor changes in our label, but I did not see 
these changes having a great impact on the treatment of patients 
with confirmed breast cancer, and also on the U.S. prevention trial, 
since the protocol did include the B-14 data, as well as the Swedish 
data, and that this potential risk was noted on the consent form." 



75 

"However, I did comment that it did have more of an impact on 
the European trial, which would require that the protocol and the 
consent form be modified. Dr. Fisher was told that Dr. Cusick had 
been notified and the European protocol and consent will be modi- 
fied. Dr. Fisher appreciated our willingness to provide him with a 
copy of the justification document for this label change. Dr. Fisher 
did comment about the potential negative publicity that could 
occur. In particular, this could be the bullet being sought by the 
health industry in the U.K. to stop the European prevention trial. 
If this is the case, that would have a major effect in the United 
States. He agreed that we should be prepared for this potential 
negative outcome." 

Now, Dr. Broder, are you aware of this exchange between Zeneca 
Pharmaceuticals and Dr. Fisher or the fact that the people in these 
studies were concerned about how the B-14 trial would effect the 
prevention trial? 

Mr. Broder. I was not aware. 

Mr. DiNGELL. Beg your pardon? 

Mr. Broder. I am not aware. 

Mr. DiNGELL. What do you make of this memo? It's something 
that should have been brought to your attention or to the attention 
of the NCI, should it not? 

Mr. Broder. I agree. I would need to consult with Dr. Ford. May 
I just take this document to Dr. Ford? 

Mr. DiNGELL. Sure. Without objection, these documents relative 
to Zeneca Pharmaceuticals will be inserted in the record at the ap- 
propriate place. [See p. 214.] 

Go ahead, Doctor. 

Mr. Broder. I'm quite disturbed by some of the things that you 
just read. 

Mr. DiNGELL. I think it would be unfair for me to try to compel 
you to comment, but it is a document which I think should concern 
us. Is it not? It tends to indicate that perhaps maybe the pharma- 
ceutical house and Dr. Fisher have not been sufficiently forthcom- 
ing, though, does it not? 

Mr. Broder. I'm comparatively concerned that I don't disagree 
with what you have just said. 

Mr. DiNGELL. Now, let's look at this situation. Here we've got the 
University of Pittsburgh. Now, the University of Pittsburgh solic- 
ited a million dollars from Zeneca for the endowment of a chair. 
They wound up getting $600,000. This is while the test of this par- 
ticular pharmaceutical is going on. 

Does that seem to be quite cricket? Dr. Varmus, do you want to 
comment? 

Mr. Varmus. I personally have concern about engaging in that 
kind of relationship. 

Mr. DiNGELL. I wonder. Does it pass the Aunt Minnie Sniff Test? 

Mr. Varmus. What test? I'm sorry. 

Mr. DiNGELL. If Aunt Minnie were to sniff this, what would she 
say? 

Mr. Varmus. Can you explain the test to me, sir? 

Mr. DiNGELL. Well, Aunt Minnie is somebody we use around here 
because she has a sensitive nose. What we're trying to figure out 
is would she like the smell of this or not. 



76 

Mr. Varmus. Probably not. 

Mr. DiNGELL. Here we've got the University of Pittsburgh. Let's 
go through some of the times that we've cut their track, just see. 
We had Dr. Steven Bruening. He pled guilty to false statements in 
connection with the making of a number of tests with regard to 
youngsters, hyperactive youngsters and mentally-impaired young- 
sters, and he said that medication should not be given them to sup- 
press the activity. He did this on the basis of studies which he said 
he had made, which, in fact, he did not make. 

Then we had a fellow by the name of Dr. Herbert Needleman. 
He studied the exposure of children to lead. Now, this was re- 
viewed by the reviewing authorities and they said that the reports 
were false, but they said that there was no misconduct. 

Then we had a fellow by the name of Dr. Charles Bluestone. He 
took large sums of money from a company that was selling drugs 
that he was studying. I wonder if we ought not get the University 
of Pittsburgh in here to talk to us about these matters. 

Mr. Varmus. I have inquired myself of the University of Pitts- 
burgh about these matters, sir. 

Mr. Broder. We have received correspondence from them on 
these points. 

Mr. DiNGELL. What do you have to comment either with regard 
to the correspondence or 

Mr. Varmus. I first learned from a letter addressed to the Sec- 
retary from Senator Rockefeller and Mr. Waxman, I believe, that 
there was a report in Who's Who that included Dr. Fisher's entry 
claiming that he occupied the ICI professorship. 

I asked Dr. Chabner, who was then in Pittsburgh, to obtain for 
me some information about this relationship, which seemed to me 
not to pass my own sniff test. What we learned was that the chair 
had been partly endowed by the ICI and that Dr. Fisher had not 
occupied that chair. Now, we looked ourselves in American Men 
and Women of Science and found an entry that included that pro- 
fessorship under his name. 

I further inquired of the university about what seemed to be dis- 
crepancies in these accounts and I was told that was a clerical 
error and that a secretary had, to Dr. Fisher's regret, included this 
in his biographical listing, that it was not accurate. 

I, myself, think that it is difficult to maintain the appearance of 
propriety and possibly the practice of propriety if one's own depart- 
ment is receiving a large endowment for a professorship from a 
company that's supplying a drug that's being used in a clinical 
study being carried out by that investigator. 

Universities are under financial stress, but the issue of the credi- 
bility of research that results from university activities is a greater 
issue, in my own mind. 

Mr. DiNGELL. I gather Dr. Fisher is still being compensated by 
the university. 

Mr. Varmus. Yes, he is. 

Mr. DiNGELL. So whether he occupies the chair is really of lim- 
ited importance. The question here is the money, not who occupies 
the chair. Somebody else is occupying the chair, but chairs seem to 
be somewhat fungible. In other words, one fellow can sit in them 



77 

or another fellow can sit in them. It doesn't really make too much 
difference as long as somebody plays the chair. 

Mr. Varmus. I'm just providing you with the answers I received. 

Mr. DiNGELL. Did you say nobody occupies the chair? 

Mr. Varmus. Apparently, the amount of money is insufficient to 
generate the income that would be required to pay a full salary. 
Now, at some institutions, I know that a chair may not pay the full 
salary, but still constitute a chair. It's a matter of definition. 

Mr. DiNGELL. I have seen nothing that would indicate on my part 
any concern about the university being distressed about this situa- 
tion. They still got the money, right? 

Mr. Varmus. Yes. 

Mr. DiNGELL. We have, then, a question about the status of the 
University of Pittsburgh's assurances. What is the status of the 
University of Pittsburgh's assurances that the — at your operation, 
if you please? 

Mr. Chabner. I'm sorry, sir. I don't understand the question. 

Mr. DiNGELL. The question was for Dr. Bivens. But, again, we 
are blessed with our inadequate 

Mr. BrvENS. An important part of OKI's work is to review the 
policies and procedures that are used by institutions, what policies 
and procedures are in place to satisfy the regulatory requirement 
for an assurance, and to see if the institution is following its own 
policies and procedures. 

We do have a major compliance review underway right now, re- 
viewing University of Pittsburgh and its compliance with the PHS 
regulations. We are first looking at their existing policies and pro- 
cedures to see if they comport with the requirements of the regula- 
tion and then we are looking at the historical record of how they 
have handled inquiries and investigations to see if they fit with 
their own policies and procedures. 

If we identify any problems, we will require immediate corrective 
action. 

Mr. DiNGELL. The assurance that we're referring to is, of course, 
the assurance of scientific integrity, isn't that right? 

Mr. BrvENS. That's correct, sir. 

Mr. DiNGELL. What happened between ORI and the University 
of Pittsburgh regarding this assurance? 

Mr. Bivens. I don't recall in my tenure any specific interchange 
between ORI and Pittsburgh related to their assurance. 

Mr. DiNGELL. Should you consider suspending this and didn't, in 
fact, you consider suspending the scientific integrity assurance 
pending improved performance by the university? 

Mr. Bivens. Not while I was Director. I became Director in Janu- 
ary 1993. There mav have been a prior history with the old offices, 
OSI and OSIR, and even my old office, OSIR, I don't recall that 
kind of interchange. But we certainly have that option of suspend- 
ing the assurance. 

Mr. DiNGELL. Clearly, your predecessor had that option because 
these events occurred apparently just previous to the time that you 
assumed your current responsibilities. 

Can you tell us what the records of your agency show? 

Mr. Bivens. I would have to look at those. I'm not familiar with 
them right now. 



78 

Mr. DiNGELL. Would you do that and make that information 
available to the committee, please? 

Mr. BlVENS. I'd be glad to. 

Mr. DiNGELL. The Chair is going to recognize the gentlewoman 
from Pennsylvania. 

Ms. Margolies-Mezvinsky. Thank you, Mr. Chairman. Dr. 
Broder, for the sake of fairness, it's important to emphasize, I 
think, that Dr. Fisher has not only not been found guilty of sci- 
entific misconduct, he's not even under investigation. Is that cor- 
rect? 

Mr. Broder. That is absolutely correct. 

Ms. Margolies-Mezvinsky. You have removed him as a prin- 
cipal investigator, as administrator of the NSABP program. We 
have been told that he was removed because NCI had some doubt 
about his continued fitness to serve as principal investigator. We've 
gotten some calls in our office from some of his patients who say 
that he was a really fine physician, saved their lives, all those 
things. 

Can you describe for the subcommittee the principal of fitness 
and how you applied it in his case? 

Mr. Broder. Thank you for the question. The word "fitness", if 
that's the right calculus that we would use here, does not go to 

Ms. Margolies-Mezvinsky. What word would you use? 

Mr. Broder. I would use suitability. But I didn't do a legal anal- 
ysis and I don't think we did the kind of exact word. But, basically, 
the bottom line is that what we felt was that Dr. Fisher could not 
be the individual with whom we corresponded on matters of the 
performance of this grant, which had to ao with things such as spe- 
cific compliance with our rules, specific implementation of auditing 
procedures, proper notification of problems, all the things that we 
have talked about and many of the issues that I raised in my open- 
ing statement. 

This was not a determination and should not be construed as a 
determination that Dr. Fisher is not fit to function as a surgeon or 
as a clinical scientist. That is not what we are saying and that is 
not what we are talking about today. That is why I tried in my 
opening statement to draw a distinction between his formidable in- 
tellect and formidable record in spite of all the things that we 
talked about, his contributions. Dr. Fisher is a Lasker Award win- 
ner. He has made a number of contributions. 

My position is none of that matters in our analysis of who shall 
run a grant that we administer. It has to do with the performance 
of the individual here and now. 

Ms. Margolies-Mezvinsky. Dr. Varmus? 

Mr. Varmus. Yes. I agree with that statement. 

Ms. Margolies-Mezvinsky. And you support the NCI's action re- 
garding Dr. Fisher. 

Mr. Varmus. The NCI recommended to the university that Dr. 
Fisher be replaced as the director of that project for administrative 
cause, and I agree with that assessment. 

Ms. Margolies-Mezvinsky. How do you view this standard of, 
whatever you want to call it, fitness, suitability, and how do you 
believe it should be applied in other cases? 

Mr. Broder. To whom is the question? 



79 

Ms. Margolies-Mezvinsky. Dr. Varmus. 

Mr. Varmus. This is a case-by-case analysis. It's an unusual cir- 
cumstance, but we had a very significant problem on our hands, 
which is the focus of this hearing today; namely, that the precipi- 
tating feature here was the failure of the NSABP to publicly dis- 
tribute the results of the ORI investigation and the reanalysis of 
the NSABP study, known as B-06. Under those circumstances, it 
was proper that the NCI evaluate administrative practices at the 
NSABP and what they found were a number of failures of adminis- 
trative oversight that, to my mind, called for Dr. Fisher's replace- 
ment. 

Now, in similar circumstances, I would advise the same thing. 
This is an unusual occasion in the history of clinical research in 
this country. At the moment, I don't see any clear simple prescrip- 
tion for the conditions under which similar actions would be ad- 
vised. 

Ms. Margolies-Mezvinsky. We are talking about the application 
of a principal, correct? 

Mr. Varmus. Yes. 

Ms. Margolies-Mezvinsky. Are we applying the application of 
this principal consistently? 

Mr. Varmus. We are, but we don't have very many cases to apply 
it to at this point. 

Ms. Margolies-Mezvinsky. Dr. Broder, does the principal of fit- 
ness or suitability also apply to intramural NCI scientists? 

Mr. Broder. It most certainly does. 

Ms. Margolies-Mezvinsky. What about Dr. Gallo? 

Mr. Broder. Is there a specific question about Dr. Gallo or do 
you wish me just to make a general comment? 

Ms. Margolies-Mezvinsky. Is it still under consideration or are 
you applying that principal? 

Mr. Broder. Dr. Varmus and I have discussed a number of is- 
sues related to Dr. Gallo. The issues involving Dr. Gallo have until 
recently been complicated by a formal inquiry process that has 
gone from OSI to ORI. We were awaiting the results of that process 
to end and also to have the kind of factfinding that we needed to 
do in order to make a decision. 

In Dr. Fisher's case, we believed that there was sufficient and 
compelling facts at our disposal, specific facts that related to his 
specific performance in a specific way, including, among all the 
other things, our detection of a report of yet an additional site in 
Canada which subsequently turned out to be an indicia of serious 
data manipulation in another patient, which Dr. Fisher had still, 
after all of our warnings, still not reported to us. 

So we believed that the facts in that case could allow but only 
one conclusion in this specific situation. But we will review the sit- 
uation and the facts of Dr. Gallo's case. Staff members have been 
very kind in providing information to me and I believe will be 
meeting with me again. 

Ms. Margolies-Mezvinsky. I don't think we're trying to get into 
the personnel action. I think we're trying to get into the principal, 
if you're going to apply the same principal. 

Mr. Broder. The answer is yes. We will apply it to all grantees 
and to the intramural program. But the jobs may be different. 



80 

There may be different aspects of what you apply to each different 
employee. Dr. Fisher has the right to assert to us that he's a very 
good surgeon and a very good clinical researcher and were we deal- 
ing with a grant that uniquely was funding him as a surgeon or 
as a clinical researcher or as someone who took care of patients, 
as you may have implied, we would not be saying we have concerns 
about his capacity to provide excellent state-of-the-art care to a pa- 
tient. That is not the basis of his removal from being a principal 
investigator. 

Ms. Margolies-Mezvinsky. Do you believe that the principal of 
fitness, suitability or whatever you want to call it should apply to 
all NIH intramural scientists? 

Mr. Varmus. Yes, I do. 

Ms. Margolies-Mezvinsky. Dr. Broder, isn't it true that NCI 
has always possessed the authority and the ability to apply these 
standards and to take the actions that you have taken in this case? 

Mr. Broder. Yes. 

Ms. Margolies-Mezvinsky. Have these standards and actions 
previously been routinely applied or are we seeing something new? 

Dr. Broder. I am not aware of — I personally am not aware, and 
staff will have to inform me. I am not aware of the NCI asserting 
its rights under the Code of Federal Regulation and other applica- 
ble statutes to demand data from an investigator, disseminate that 
date, publish the data without the investigator's necessary involve- 
ment even, possibly even against the will of the investigator. I am 
not aware of that principal ever having been implemented. But it 
has been now and we will not hesitate to do it again if a fraud 
issue comes up. 

It has a downside, however, in that it could create situations in 
which we at NCI are making statements which are at variance 
from a formidable intellect and leader in a field, and that also has 
its downsides. That's why our hope had been — this is an expla- 
nation, not an excuse. 

Our hope had been that we could reach a situation where Dr. 
Fisher, with our assistance, would take the lead to first come for- 
ward with all the different issues. 

Ms. Margolies-Mezvinsky. Dr. Varmus, do you support NCI's 
willingness to exercise this existing authority and will we see simi- 
lar actions, if necessary, across the board at NIH? 

Mr. Varmus. I support it in this case and, by implication, I 
would support it in other situations that are comparable. 

Ms. Margolies-Mezvinsky. Thank you. I yield the balance of my 
time, Mr. Chairman. 

Mr. DiNGELL. The Chair recognizes the gentleman from Ohio. 

Mr. Brown. Thank you, Mr. Chairman. Dr. Broder, on the sub- 
ject of willingness to deal forthrightly with scientific fraud in, I be- 
lieve, July of 1991, you were briefed by OSI concerning its findings 
of falsification and its findings of fabrication of St. Luc data. 

Describe, if you would, what you were told, what directives you 
issued pursuant to that briefing. 

Mr. Broder. Yes. I believe the committee has a memo for the 
record prepared by Dr. McFarland in this regard. I received a sum- 
mary from individuals with firsthand knowledge of the fraud at Le 
Hopital St. Luc in Montreal. It was very clear to me that there was 



81 

a serious problem of overt fraud and that the issue had to be dealt 
with forthrightly and effectively. 

We determined that we would cooperate with ORI and provide 
whatever resources we could. I determined at that time that the 
only course of action that could be taken would be to segregate the 
data from the Le Hopital St. Luc and permanently pull it out of 
future analyses, but providing the data — providing analyses both 
with and without the data, but permanently notifying individuals 
that the data had to be pulled out, and, also, taking steps to appro- 
priately reanalyze and republish the data with the appropriate 
interactions with the Office of ORI, although it may not have been 
called ORI in that era. 

It was at that point the predecessor organization for ORI. 

Mr. Brown. So I take it NCI and Pittsburgh simply didn't follow 
your direction. 

Mr. Broder. Speaking as to the effects of the University of Pitts- 
burgh, they certainly did not follow our very constructive criti- 
cisms. As to NCI staff, I will accept responsibility for this issue. 

Mr. Brown. This seems to be the picture. We have the Director 
of NCI telling subordinates to have University of Pittsburgh reana- 
lyze the data, republish the analysis, not publish further studies 
using this falsified fabricated data; yet, every single directive was 
disobeyed, ignored, taken too lightly, whatever. Why? What hap- 
pened? 

Mr. Broder. I believe it is, in part, a function of Dr. Fisher's for- 
midable reputation and I believe that the staff were attempting to 
negotiate a collegial non-confrontation way of dealing with a pio- 
neering figure who obviously knew a great deal. I believe there was 
an excessive level of collegiality and a higher level of tolerance 
than is now the case. That is probably the best way that I can sum- 
marize this. 

Staff simply did not wish to confront and order Dr. Fisher, who, 
after all, is the person that had made many contributions, had a 
great deal of status in the scientific community. I believe that is, 
in part, responsible for what happened. 

The other issue, which is of no comfort to the committee or the 
public, but which is, I think, one of the factors that should be put 
on the table, is that the staff felt that Dr. Fisher had been right 
on a number of occasions that, in fact, there were no changes that 
would come from this and that other studies already available were 
confirming the value of breast-sparing surgery. So all of those fac- 
tors came into the mix. 

I do believe that had there been a public health hazard, that the 
other steps would have been taken, but that's of no consolation to 
the committee and I accept your point. 

Mr. Brown. Dr. Varmus, according to Dr. Broder, it sounds a lit- 
tle bit like Congress, collegiality, protecting people, people protect- 
ing themselves, all of that. It sounds like the criticisms that people, 
sometimes rightly, level at this institution. 

What is the rationale beyond that for why Dr. Fisher and his col- 
leagues would continue to submit and publish papers that are 
known to contain that false data? What, building on what Dr. 
Broder said, is that rationale for Dr. Fisher to do that? 



82 

Mr. Varmus. I think these are questions that should be ad- 
dressed to Dr. Fisher. But I think there are some potential expla- 
nations that have to do with the desire to publish the findings of 
studies that have been carried out in multiple-institutions. 

I cannot myself condone the inclusion of data from the St. Luc 
Hospital once fraud at St. Luc's had been ascertained. To my way 
of thinking, information from that hospital should have been ex- 
cluded from any further publication. So I don't want to pretend to 
understand how that came to be. I think you really need to address 
those questions to Dr. Fisher and other members of the NSABP. 

Mr. Brown. Let me ask it a different way. Let me ask Dr. Broder 
in this case. The subcommittee staff has found that as early as 
July of 1992 that NCI officials were admonishing Dr. Fisher, ad- 
monishing his colleagues to upgrade and strengthen auditing proce- 
dures. What was the response of Dr. Fisher and what was the re- 
sponse of his colleagues to those repeated requests from NCI? 

Mr. Broder. I would say that Dr. Fisher's response to us was 
quite disrespectful of the role that government employees play and 
quite disrespectful of the status and functions that we have and, 
I think I'm accurately paraphrasing, basically said words to the ef- 
fect of who are you to criticize me, I know how to do clinical trials, 
I've been doing them before you were a doctor. 

That's not literally what he said, but I am giving you my edi- 
torial response. I believe our staff referred to some of his things 
and tried to give him constructive criticism, which he rejected out 
of hand. 

Mr. Brown. So what's the next step? What was the next step? 
If he responded to you that arrogantly and with that haughtiness, 
if you will, and with that self-certitude or self-righteousness, what 
was your response back? 

Mr. Broder. I believe the staff continued the process of negotiat- 
ing with him and bending him to the things that needed to be 
done. They did, in fact, negotiate successfully for him to accept cer- 
tain changes, at least as we were informed about them. We made 
our point with respect to the institution of data safety monitoring 
boards and other things that he was instructed to do. 

They had been given assurances that papers were being prepared 
and that reanalyses were coming, and I think that that — it was a 
slow process, but I believe, speaking for the staff, they thought 
they were moving in the right direction and that they were pre- 
serving Dr. Fisher as a force who could contribute and continue to 
do good in a method that was collegial. 

Mr. Brown. We have seen a number of reports of NSABP audits 
that reflect significant discrepancies in the audited records of var- 
ious research sites. What are NSABP's duties and responsibilities 
regarding audit follow-ups and recommendations? 

Mr. Friedman. Currently, under the new branch which has been 
just identified and even prior to the establishment of that branch, 
the NSABP is having a complete overhaul of how their on-site au- 
diting is conducted. If you're interested, I can describe the inad- 
equacies of the previous program or I can simply give you informa- 
tion about the current program. 

Mr. Brown. Do both, please. 



83 

Mr. Friedman. All right, sir. In the past, rather than having an 
audit where investigators would go to the site that the research is 
being conducted and to examine records from that site and look for 
supporting information, the way the NSABP conducted their audits 
was to have individusds go xerox records and bring them back to 
Pittsburgh. In addition, whereas our other cooperative groups — and 
this is a very important point, sir. 

In contrast to how our other cooperative groups operate, the 
number of charts sampled at each of these research sites was rel- 
atively small. Instead of having a larger number of charts sampled 
from a larger enrolling institution, they had a small fixed number 
of charts examined at each institution. 

This was not — although it was defended by the NSABP, this was 
not a system that we felt was entirely appropriate or would be as 
efficient as we would like. We were very concerned about the dis- 
covery of the fraud at St. Luc Hospital and wanted very much to 
try and look carefully at their system and bring it at least up to 
the standards of other cooperative groups. 

Currently, the monitoring program for the NSABP is to have a 
larger number of charts sampled, to have researchers go to those 
individual sites and actually look at the primary data, an x-ray or 
an EKG form or whatever is at that site, to confirm the reliability 
and truthfulness of that. 

We have been very concerned about the whole credibility of the 
lumpectomy trial because, as was expressed this morning, there is 
so much public concern about it and it would be wrong for us to 
not be attentive to that concern. We have been very careful to go 
and look at more than 850 research records, the primary patient 
charts of patients who enrolled in the B-06 — that is the 
lumpectomy trial — to examine those records to see if there's any 
systematic fraud or falsification. 

We also examined a number of other charts, all together almost 
1,400 charts, looking at many different studies, to try and confirm 
that, in fact, despite the unsatisfactory nature of the system that 
existed previously, that we could still have confidence in the con- 
clusions that these investigators were describing. 

Mr. Brown. Thank you, Mr. Chairman. 

Mr. DiNGELL. The Chair thanks the gentleman. The gentleman 
from Colorado. 

Mr. SCHAEFER. Thank you, Mr. Chairman. Just a few wind-up 
questions. Dr. Varmus, I understand that approximately a million 
dollars of the taxpayers' money was used in the St. Luc research. 
Are there any efforts underway to determine whether this money 
can be reclaimed? 

Mr. Varmus. Yes, there are. We've been trying to reclaim the 
money. As you understand, there are jurisdictional problems, with 
the St. Luc Hospital being located in Montreal, and we've asked the 
Department of Justice for help in trying to recover the funds. 

Mr. SCHAEFER. Does the University of Pittsburgh — after you. 

Mr. DiNGELL. Would you give us a report on the status of mat- 
ters with regard to the assistance of the Department of Justice, 
please? 

Mr. Varmus. For the record, yes. 



84 

Mr. DiNGELL. Not right at this minute, but just when it's com- 
fortable. 

Mr. Varmus. Absolutely. 

Mr. SCHAEFER. May the University of Pittsburgh have any liabil- 
ity in this? 

Mr. Varmus. They might. We're looking into that as well. 

Mr. SCHAEFER. That's all part of it. And this is what we're going 
to get an update on. 

Mr. Varmus. We can provide you with a detailed account of that 
from our general counsel. 

Mr. ScHAEFER. Can you tell me if these efforts to retrieve the 
money were begun before or after the subcommittee started looking 
at this? 

Mr. Varmus. As far as I know, after. 

Mr. SCHAEFER. After. 

Mr. Varmus. After I was informed of some of these concerns. 

Mr. SCHAEFER. Right. 

Mr. Varmus. As you know, I'm new at NIH, but when I heard 
about them, we began to look into this question. 

Mr. SCHAEFER. So once our subcommittee began looking into this 
matter, then steps were taken to try to reclaim it. 

Mr. Varmus. That's correct. 

Mr. SCHAEFER. I thank the subcommittee for that. It's a million 
dollars. Are you looking in your files to see if there's attempts 
being made to recover taxpayers' dollars in other instances where 
fraud was involved? 

Mr. Broder. Sir, we are. We consider that we — we have a man- 
ual that has been implemented in which fraud will always be pre- 
sumed, a rebuttable presumption, but will always be presumed to 
indicate recovery. Then we will act against the grantee according 
to the advice that we receive from the Office of General Counsel 
and according to applicable principals of law and Federal regula- 
tions. 

This will be automatic and it is clearly understood in our grants 
management process and so on. 

Mr. SCHAEFER. Had this been done before this situation, before 
our subcommittee started 

Mr. Varmus. I've been informed by Dr. Baldwin it's been done 
twice before. 

Mr. SCHAEFER. OK. Mr. Chairman, I'm finished. 

Mr. DiNGELL. The Chair thanks the gentleman. Dr. Bivens, what 
other cases presently active at ORI or other during your tenure 
that have been closed with a finding of misconduct involved clinical 
trials or impact upon the public health? 

Mr. Bivens. I believe that the St. Luc case is the only one in 
which there was a confirmed scientific misconduct finding related 
to a clinical trial. 



85 

Mr. DiNGELL. Which one would that be? Can you tell us about 
presently active ones? 

Mr. BiVENS. Presently active ones? 

Mr. DiNGELL. Yes, sir. 

Mr. BrvENS. My understanding is that there is an investigation 
underway and an inquiry underway on two trials funded by the 
National Eye Institute. 

Mr. DiNGELL. National Art Institute? 

Mr. BlVENS. Eye Institute. I'll be glad to provide more informa- 
tion on those. I don't have it ready at hand at the moment. 

Mr. DiNGELL. If you please. 

[The following information was received:] 



86 



REPORT ON ACTIVE ORI CASES INVOLVIKG MXILTICENTER CLINICAL TRIALS 

A. Investigation into possible falsification and fabrication of 
data in the PES funded COLLABORATIVE OCULAR MELANOMA STUDY 
at Northwestern University Medical Center (93-27) and The 
Cleveland Clinic Foiindation (93-28) 

Funded by the National Eye Institute (HEX) , National Institutes 
of Health, The Collaborative Ocular Melanoma Study (COMS) 
consists of two multicenter, randomized controlled clinical 
trials designed to investigate the efficacy of radiation therapy 
compared to surgery in prolonging the vision and survival of 
patients presenting with choroidal melanoma, a rare intraocular 
cancer . 

Approximately 1500 new cases of choroidal melanoma are diagnosed 
each year within the U.S. The liver is the most frequent site of 
metastasis. The median survival after diagnosis of metastatic 
disease is 6-9 months; therefore, choroidal melanoma is indeed 
life threatening. The most widely used treatment for choroidal 
melanoma of all sizes has been removal of the eye (e.g., 
enucleation) . Radiation therapy has been advocated as an 
alternative treatment with the goal of preserving the eye and 
possibly some vision in the affected eye. 

Patients presenting with unilateral choroidal melanoma classified 
as "medium" in size are randomized with equal probability to 
either enucleation or radiation therapy by application of a 
radioactive plaque sutured to the eye over the base of the tumor. 
Patients with large tvimors are randomized with equal probability 
to either enucleation or a five day course of external beam 
radiation therapy followed by enucleation. ( Controlled Clinical 
Trials . 14:362-391, 1993). 

Clinical data on COMS-eligible patients in each of the more than 
4 participating centers is forwarded to the COMS Coordinating 
Center for entry into the study database and statistical 
analyses. The Coordinating Center is located at the Wilmer 
Ophthalmological Institute, The Johns Hopkins University, 
Baltimore, MD, and is supported by a cooperative agreement. 

No results of the COMS clinical trials have been published at 
this time. The protocol for the study was published in 1993 and 
is referenced above . 

The COMS Coordinating Center identified possible falsification 
and fabrication of records submitted from two of the 
participating COMS centers, the Northwestern University Medical 
Center and The Cleveland Clinic Foundation, on April 14, 1993, 
and on June 28, 1993, respectively. The NEI program officer was 
alerted about the discrepancies in the Northwestern COMS data on 



87 



May 21, 1993. Similar discrepancies in data from the Cleveland 
Clinic Foundation program were revealed at a meeting of the COMS 
Data Safety and Monitoring Committee on October 4, 1993. 

ORI was informed of the data reporting discrepancies at the 
Northwestern COMS center on August 26, 1993, by a telephone call 
from the NET misconduct policy officer. ORI was informed of the 
data reporting discrepancies associated with The Cleveland Clinic 
Foundation COMS center on October 5, 1993, during a meeting 
between NEI and ORI staff which had been scheduled to discuss 
the Northwestern findings. 

The Northwestern University Medical Center; 

The Northwestern University Medical Center has been a participant 
in the COMS program since 198 6. Until recently, the Northwestern 
COMS program registered the largest patient enrollment (103) of 
all the participating COMS centers nationwide and in Canada. In 
1992-1993, Northwestern University received $127,700 in PHS 
support for its participation in COMS. 

Identified discrepancies in the reporting of clinical trial data 
at the Northwestern COMS Center were brought to the attention of 
University officials and to the COMS Coordinating Center on April 
14, 1993, by the Principal Investigator. He stated that 
inaccurate data (e.g., misrepresentation of examination dates and 
other dates associated with laboratory tests. X-rays, etc.) were 
supplied to the COMS Coordinating Center by the Northwestern COMS 
clinical coordinator who was immediately placed on leave from the 
progreua; employment was terminated later in April without an 
official University inquiry. The COMS Coordinating Center 
dispatched an audit team to the Northwestern COMS center on April 
22 and again on June 1, 1993. A University committee of inquiry 
was established on May 20, 1993, to examine the issue of possible 
scientific misconduct within the COMS program and to make a 
recommendation to University officials regarding the need for a 
full investigation. The inquiry committee submitted its 
recommendation that an investigation was warranted in a report to 
the Vice President for Research and Graduate Studies on October 
5, 1993. 

The Division of Research Investigations (DRI) in ORI was notified 
on August 26, 1993, by the NEI of data reporting discrepancies at 
the Northwestern COMS center and told that the institution was 
conducting an inquiry. DRI met with the NEI misconduct officer 
to obtain additional information and a copy of the COMS 
Coordinating Center audit report, and informed Northwestern 
University officials on October 12, 1993, of its intent to open a 
case and to conduct a direct investigation. The DRI requested 
the University to cease its investigation of the COMS matter and 
to secure all COMS patient medical and research files. Based on 
the Coordinating Center audit report and the Northwestern inquiry 
report, DRI identified the potential respondents in the case. 



88 



DRI visited Northwestern University on February 7-11, 1994. 
Medical and research files of all 103 eligible patients enrolled 
in the COMS program were reviewed against data submitted to the 
coordinating center. COMS Coordinating Center staff provided 
technical support during the review. Several key witnesses and 
all but one of the possible respondents were questioned and their 
testimony recorded. The DRI analyses of the evidence obtained at 
the site visit and final conclusions remain to be finalized at 
this time. 

There were a number of data items reported for the Northwestern 
patients which could not be confirmed because documentation was 
not available. These issues remain to be resolved. 



The Cleveland Clinic Foiuidation: 

The Cleveland Clinic Foundation has been a participant in the 
COMS program since 198 6; investigators have entered 24 patients 
on the study. The Cleveland Clinic Foundation receives funds 
through a subcontractual agreement with the COMS Coordinating 
Center at the Johns Hopkins University; funds allocated to The 
Cleveland Clinic Foundation COMS program for the ninth year 
(1993-1994) of the study total $20,762. 

Identified discrepancies in the reporting of clinical trial data 
from The Cleveland Clinic Foundation came to the attention of the 
COMS Coordinating Center through a routine audit of patient files 
by a representative of the Coordinating Center on June 28, 1993. 
The initial purpose of the visit by the Coordinating Center 
representative was to review the current status of clinic 
operations and to assist in resolving any problems that arose 
since the previous clinic visit or that were identified during 
the current visit. Identified discrepancies included: 

1) blood test results were reported as normal but no 
record was available to confirm the test was performed; 

2) falsification of dates for patient examinations so as 
to be compatible with protocol restrictions; 

3) reporting of blood test results as normal when the 
results were abnormal; 

4) a COMS certified examiner was identified as having 
performed an evaluation when the evaluation was 
actually performed by a non-COMS certified staff 
member. 

Although the data reporting discrepancies associated with The 
Cleveland Clinic Foundation COMS center were known to the COMS 
Coordinating Center at the end of June 1993, this information was 
not brought to the attention of the NEI until the October 4, 
1993, meeting of the COMS Data Safety and Monitoring Committee. 



89 



The NEI Program Officer alerted the NEI misconduct policy 
officer, on October 4, who, in turn, alerted the ORI on October 
5, 1993. 

On October 12, 1993, DRI informed institutional officials at The 
Cleveland Clinic Foundation of its intent to open a case (DRI 93- 
28) and to conduct a direct investigation into the alleged 
scientific misconduct involving The Cleveland Clinic Foundation 
COMS program. The institution was instructed not to initiate its 
own inquiry or investigation and to secure all research and 
medical records of every patient affiliated with the COMS 
progreim. Based on the Coordinating Center report and their 
positions with regard to the COMS project, DRI named the possible 
respondents in the matter. 

The DRI site visit to The Cleveland Clinic Foundation COMS center 
was conducted on December 6-9, 1993. Medical and research files 
of the 24 eligible patients enrolled in The Cleveland Clinic 
Foundation COMS program and of 37 patients screened but not 
eligible for the COMS program were evaluated for possible 
fabrication and/or falsification of reported data. COMS 
Coordinating Center staff provided technical support during this 
review. Key witnesses and all respondents were interviewed and 
their testimony recorded. 

The DRI has concluded its fact-finding of The Cleveland Clinic 
Foundation COMS program and is preparing its final report. 

ORI has briefed the Office of the Inspector General, HHS on DRI 
#93-27 and #93-28. Additionally, ORI met with officials from NEI 
on April 19, 1994 to brief them on the status of the cases. At 
that meeting,. ORI and NEI staff agreed that a letter would be 
sent by NEI to all COMS patients regarding the investigations. 
NEI confirmed that this letter had been sent to every patient by 
April 29. NEI has also prepared a statement which will be 
published in the May 18 issue of the Journal of the American 
Medical Association . 

Potential compliance issues concerning the reporting of 
scientific misconduct to the ORI and the termination of 
employment prior to an inquiry/ investigation remain to be 
assessed. 

B. Inquiry into possible falsification and fabrication of data 
in the PES funded ISCHEMIC OPTIC NEUROPATHY DECOMPRESSION 
TRIAL at Ohio State University (DRI 94-04) 

The Ischemic Optic Neuropathy Decompression Trial (lONDT) is 
another multicenter clinical trial supported by the NEI. The 
purpose of the trial is to assess the safety and efficacy of 
optic nerve sheath decompression surgery for non-arteric ischemic 
optic neuropathy (NAION) . NAION is the most common cause of 
acute optic nerve disease in older persons and causes permanent 



90 



and severe visual loss. It is estimated that 6,100 new cases of 
NAION are seen each year. The lONDT is a randomized clinical 
trial designed to compare the difference in change in visual 
acuity at six months in patients assigned to either surgery or 
careful followup. Begun in November 1992, the trial has 26 
participating centers. 

On Wednesday April 6, 1994, the Director, lONDT Coordinating 
Center, University of Maryland at Baltimore, notified the NEI 
program officer and misconduct officer and also called the 
Division of Research Investigations (DRI) to report possible 
misrepresentations of information concerning subjects in the 
trial provided to the Coordinating Center by members of the 
project team at Ohio State University (OSU) . The lONDT project 
at OSU has received funding from PHS under a cooperative 
agreement . 

During a December 1993 site visit, numerous protocol violations 
had been observed and a decision was made to closely monitor the 
OSU contributions. During the recent site visit (April 4, 1994), 
the coordinating center staff uncovered discrepancies in 
infonnation that was provided regarding randomization of subject 
027-15 and about subjects' visual acuity measurements at 
randomization (baseline) which raised the possibility that data 
had been falsified or fabricated. 

From information sent from the Coordinating Center on April 8, 
1994, concerning the visual acuity measurements at baseline for 
some of the OSU patients, DRI indicated to OSU officials that an 
inquiry would likely be necessary. The Coordinating Center sent 
DRI a copy of a draft site visit report on the evening of Friday, 
April 8. 

The specific allegations of scientific misconduct as formulated 
by DRI are that: 

1) an individual or individuals at OSU falsified or 
fabricated information provided to the lONDT 
Coordinating Center; 

2) falsified information regarding randomization of 
subject #027-15 was reported verbally to Coordinating 
Center staff during the April 4, 1994, site visit; 

3) visual acuity scores of patients randomized to either 
careful watching or surgery were fabricated or 
falsified. 

DRI requested that OSU conduct an inquiry on April 11, 1994. 
University of Maryland officials were also notified of this 
decision because the continuing cooperation of Coordinating 
Center staff in providing materials, information and possibly 
technical support would be required. 



91 

Mr. DiNGELL. Grentlemen, we've kept you here a long time. I 
want to say a couple things. The NIH is a national treasure and 
you are guardians of a national treasure. It's one that this commit- 
tee and the chairman of this subcommittee and this subcommittee, 
my good friend, Mr. Schaefer, have most actively supported and we 
intend to continue doing that. 

You have had a long and a difficult appearance before the com- 
mittee. I want to make it very clear to you, Dr. Varmus, and to 
you. Dr. Broder, that I have immense respect for you both. And I 
want to make it very clear that I think that the comments that you 
have made today with regard to the attitude and the behavior of 
NIH and NCI with regard to scientific misconduct, the protection 
of participants in trials and tests and research studies that are 
funded by your agency have been most helpful and are important. 

I want you to know that we've long been critical of science and 
there's been a long effort on the part of this subcommittee to see 
to it that government money is properly spent, that the protection 
of participants in trials and tests and studies is adequate, the in- 
formation available to them is at a level that you and I would like 
to see them have, and, also, to see to it that the public moneys are 
well spent and that the factual results of the studies are factually 
reported, and that there's not misbehavior in terms of fraudulent 
studies or falsification of data or information and so forth. 

It's been a long and hard and difficult task. We found today a 
new component which is now very clear, and that is it's not just 
an isolated question in which science will rectify the falsification or 
the improprieties with regard to data and collection of information, 
but, rather, that human lives are at stake. 

We sense that there is a great awareness on your part, Dr. 
Varmus and Dr. Broder. I am content to continue to work with you 
and to watch and see to it that the kind of effort that you are mak- 
ing, which is significantly different and better than your prede- 
cessors, including one of your most immediate predecessors. Dr. 
Varmus, and to see to it that you succeed in your efforts to make 
this system work better for the benefit of all of us, that the tax- 
payers' money be better spent, that the protection be afforded to 
the participants, that the information be such as to give an honest 
appraisal so that people can make proper judgments with regard 
to their personal activities, as Ms. Sigal has indicted to us. 

So as we close the hearing, I want to express my thanks to you 
both and also to Dr. Chabner, Dr. Friedman, Dr. Bivens, for your 
assistance. You've got over there a very nasty job. Doctor, and as 
for you. Dr. Varmus and Dr. Broder, because occasionally you've 
got to address the question of scientific behavior within and with- 
out the institution and whether or not it conforms with what I 
think we regard as being the necessary tests. That is that science 
should be the pursuit of truth and that it should be honestly con- 
ducted, with minimal risk and hazard to those who innocently par- 
ticipate in these studies, and that the factual results should be that 
which add to our knowledge and which increase the safety and the 
protection of those who pay for it, and that the public money 
should be properly spent. 

I think that you and I look forward to a time when we're going 
to be able to work better together. If things have happened today 



92 

which cause you distress, you have my expression of sorrow and 
apology. 

I want you to understand that we're going to try and work with 
you. We expect good things from you. We know that you were en- 
gaged in a difficult task and, as I've indicated, the communication 
to the public of potential wrongdoing of a colleague or a scientist, 
particularly one who might happen to be a famous name or well 
known and influential figure in science, is always difficult. It's 
fraught with peril to you, possible lawsuits and other difficulties. 

We understand that it takes time to refine the procedures that 
you have so that you can do this thing properly. We will work with 
you to get you both the time and a climate in which you may do 
these things with minimum peril to yourself and greater success 
from the standpoint of the great undertaking of which you are a 
part. 

So you have our thanks and our good wishes and we will look 
forward to working with you. We will also look forward to the next 
hearing in this matter, which will be occurring, I suspect, in the 
not far future involving the University of Pittsburgh. 

Having said those things, we express to you our thanks and our 
good wishes, gentlemen, and thank you for being with us today. 

The subcommittee will stand adjourned till the call of the Chair. 

[Whereupon, at 2:15 p.m., the subcommittee adjourned, to recon- 
vene at the call of the Chair.] 

[The following letter was received:] 



93 

ZENECA 



1800 Concord Pike 
Wilmington 
Delaware 19897 USA 

Telephone (302) 886 775 1 
Fax (302) 886-7688 



Phermaceuticals Group 



Alan J. Milbauer 
Vice President 
External Affairs 



April 20, 1994 

The Honorable John D. Dlngell 
Chairman, Subconnittee on Oversight and 

Investigations 
Coanittee on Energy and Commerce 
Room 2323 Raybum Hotise Office Building 
Washington, D.C. 20515 

Dear Hr. Chairman: 

I an writing respectfully to request a correction of the record concerning 
an issue which was raised in the Oversight and Investigations Subconnittee ' s 
April 13 hearing on breast cancer research. 

In the course of the Subcommittee's hearing, an internal Zeneca memorandum 
(dated July 7, 1993), which we provided to the Subcommittee, was quoted from 
and discussed. In questions you asked of Dr. Broder of the National Cancer 
Institute (NCI), it was implied that this memorandum proved that the NCI had 
not been notified in a timely way of information Zeneca had received 
concerning the number of patients within the B-I4 trial who were developing 
endometrial cancer while on NOLVADEX. 

This is simply not true. In fact, the memorandum cited in the hearing 
summarizes a conference call which was held between Dr. Leslie Ford of the 
NCI, Dr. Fisher of the NSABP and Dr. Paul Plourde of Zeneca to discuss the 
endometrial cancer issue and the need to modify the label on NOLVADEX. This 
conference call was initiated by Dr. Plourde as soon as the endometrial 
cancer information came to his attention. Furthermore, another document we 
provided to your subcommittee was a copy of a July 8, 1993 letter from Dr. 
Plourde to Dr. Fisher of the NSABP and Dr. Ford of the NCI. This letter 
discusses the findings concerning endometrial cancer and Zeneca's proposed 
amendments to the prescribing information for NOLVADEX. Finally, on June 
30, 1993, Zeneca also notified the FDA of these findings and of the proposed 
labeling changes. 

These documents are clear: As soon as Zeneca obtained information 
concerning the increased incidence of endometrial cancer in the B-14 trial, 
we alerted the NCI and made appropriate changes to the prescription 
information for NOLVADEX. In short, Zeneca acted promptly and responsibly. 

I appreciate this opportunity to correct the record concerning the 
unfortunate implication that Zeneca was somehow deficient in reporting 
information to the NCI. I respectfully request that this letter and the 
accompanying attachments be entered into the record of the April 13 hearing. 



Sinceraiy, 




Ian J 
AJM/gfi 
Attachments 



94 
i,tNtC_/\ Pharmaceuticals Group 



Internal Heinorandum ZENECA, INC. 

UILMINGTON, DE 19897 
CLINICAL & MEDICAL AFFAIRS 
ENDOCRINOLOCT 

TO: SEE ATTACHED LISTING 

DATE: JULY 7, 1993 

FROM: PAUL V. PLOURDE, M.D. CC: 

RE: CONVERSATION WITH DR. FISHER (NSABP) & DR. FORD (NCI) 

PRIVILEGE AND CONFIDENTIAL: ATrORNET-CLIENT INFORMATION 

Today on July 7, 1993, at 1:30 p.a. I called Dr. Fisher to infora hla on the 
company's position on the association of endoaetrial cancer vith NOLVADEX 
treataent. 

I first discussed vith Dr. Fisher that ve over the last fev veeks have been 
looking at the association of endoaetrial cancer vith NOLVADEX treataent. 
This vas in part initiated because of the Tale publication reporting that the 
tuaors associated vith NOLVADEX treataent vere poorly differentiated tuaors 
resulting in a poor prognosis for these patients. Our evaluation froa the 
clinical trial data base as veil as froa the literature could not confira this 
Tale publication. Therefore, ve did not feel that ve needed to aake any 
changes in our label. Hovever, viligence is necessary and ve vill continue to 
aonitor this closely. Dr. Fisher thought that this vas appropriate and he 
hiase}f has Initiated soae activity vithin the NSABP to explore this further. 
Be is requesting that slides froa the endoaetrial cancer seen in the B-14 
trial be evaluated. He plans on doing this over the next several aonths. - 

I then proceeded to tell Or. Fisher that the increasing nuaber of patients 
vithin the B-14 trial developing endoaetrial cancer vhile on NOLVADEX did 
proapt us to look at this issue aore closely. Upon careful reviev of the 
data, ve felt that there vas an Increase risk to develop endoaetrial cancer 
vith NOLVADEX and that ve vill aodify our label to reflect this. I also 
coaaented that it vas iapossible to precisely quantitate the relative risk 
although qualitatively it did appear that there vas a slight increase 
incidence. Or. Fisher agreed vith this assessaent and felt that ve vere 
acting responsible by changing our label. I coaaented to hla that at least in 
the United States, ve vould have to aake ainor changes to our label but I did 
not see these changes having a great iapact on the treataent of patients vith 
confiraed breast cancer and also on the US prevention trial since the protocol 
did include the B-14 data as veil as the Swedish data and that this potential 
risk vas noted on the consent fora. Hovever, I did coaaent that it did have 
aore of an iapact on the European trial vhlch vould require that the protocol 
and consent fora be aodified. Or. Fisher vas told, that Dr. Cusick has been 
notified and the European protocol and consent vill be aodified. 

AbwtMwu wn««rfnMC. - 
*ie*w«ei HMCCA ^^*-»* • . 



95 



CONFIDENTIAL 

Page 2 

Conversation with Or. Fisher 

Dr. Fisher appreciated our willingness to provide hin vith a copy of the 
justification document for this label change. Dr. Fisher did coinnent about 
the potential negative publicity that could occur. In particular, this could 
be the bullet being sought by the health authority in the UK to stop the 
European prevention trial. If that is the case, that vould have a major 
effect in the United States. Be agreed that ve should be prepared for this 
potential negative outcome. 

I told him that ve vould be gathering a panel of experts in endometrial cancer 
together vithin the next tvo or three months. This vould be an intematlonal 
panel composed of the prevention trial investigators, epidemiologist and 
gynecologists. Dr. Fisher did not see the great need for having this panel, 
since he felt that the issue had been examined and he sav no need to rediscuss 
this issue. Hovever, he understood our desire to convene this panel and 
agreed to participate. 

I also Informed him that ve did not, as a company, see that the risks had 
changed appreciably to varrant discontinuation of medication to the NCI. 
Dr. Fisher vas relieved vith that decision. 

I also informed him of Dr. Cuzick's plan to hold a press conference this 
coming Friday morning. He vas quite surprised about the press conference 
although he did knov that Dr. Cuzick vas having an investigators meeting that 
day. Or. Cuzick had requested an NSABP representative to attend this meeting. 
The NSABP due to time constraints, vas unable to send anyone and Dr. Cuzick 
then called Dr. Ford at the NCI and it vas agreed that Or. Susan Nayfield from 
the NCI vas to represent the US. Dr. Fisher and Dr. Ford vere unavare that a 
press conference vas going to be held. 

I told Dr. Fisher that I vanted to call the NCI and Inform him of these 
changes. He insisted that he call Dr. Ford to discuss the Issues, but finally 
he agreed to have a conference call vith Dr. Ford. 



Later this same day, a conference call vas held betveen Dr. Leslie Ford from 
the NCI and Dr. Fisher from the NSABP. Dr. Fisher had briefed Dr. Ford on the 
issues that ve had discussed above. Dr. Ford had no comments and agreed vith 
our approach in re^rds to the label change. She did not feel that, the US 
needed to make any changes in the consent form or the protocol since the 
endometrial cancer Issue that had already had been addressed consistant vith 
our nev label change. She recognized hovever, that this could cause some 
unnecessary and inappropriate publicity. Vith this. Dr. Ford did have 
concerns about Dr. Nayfield being in the UK during the press conference and 
the meeting. She vill be contacting Dr. Nayfield to alert her of our 
intention to modify our label. Or. Nayfield vill be instructed not to 
participate in the press conference. 



96 



CONFIDENTIAL 

Page 3 

Conversation vith Dr. Fisher 



Dr. Ford also told me that the MRC had recently approved the protocol this 
veek. This needs to be confiraed, but if true, this vould ainiaize the inpact 
of Cuzick's press conference, (i.e., the UK Health Authority may not be in a 
position to be embarrassed as vas originally feared). 

I discussed the panel of experts to be organized to revlev the data. Both 
£elt that this vas unnecessary. She informed me that as far the Prevention 
Trial, a sub-study vlll be conducted where endometrial sampling vill be done 
periodically during the trial. 

I told Drs. Ford and Fisher that I vould be sending them a copy of the 
Justification document and that they should hold this document as confidential 
information. I also informed them that I vould be glad to send them a copy of 
the specific wording change ve are recommending as a label change for their 
information. 

Overall, this interaction vas positive and I believe that both groups felt 
happy that ve vere keeping them informed. 

PVP:«c 



97 



ENECA 

i.O,.„l!-..l..|-ii4r^..|!|.| 



ICOO Co-icC'-c i'.- ■ 
VVilmmgion 
Delawate 19S97 U5^ 



July 8, 1993 



Bernard Fisher, H.D. 
University of Pittsburgh 
School of Hediclne 
914 Scaife Ball 
3SS0 Terrace Street 
Pittsburgh, PA 15261 

Dear Drs. Fisher and Ford: 



Leslie Ford, M.D. 
National Cancer Institute 
Room 300 

Executive Plaza North 
Bethesda, MD 20892 



Further to our telephone conversation last week, I enclose a copy of 
the report vhich ve have prepared for discussion vitb regulatory 
authorities to support our proposal to aaend the prescribing 
inforaatlon for NOLVADEX vith respect to endonetrial changes, vith 
particular reference to endonetrial cancer. Although I assuae that 
you vill vlsh to discuss the contents of the report vith your clinical 
colleagues of the NSABP and MCI, it Is a condition of this disclosure 
to you that the Groups treat the data as confidential and does not 
disclose the contents of the report to any third party. 

ZENECA Pharaaceuticals believe that the inforaation currently 
available strengthens the association betveen NOLVADEX and an 
increased incidence of endonetrial cancer, and that this does 
therefore inpact on the assessnent of the risk benefit ratio for vomen 
participating in the current prevention trials. 

Ve are therefore inforalng all 3 groups (US, Italy in addition to UK) 
of this inforaation as one of the conditions governing our agreenent 
to supply clinical trial aedication i.e. that ve vould infora the 
groups of any significant inforaation relevant to tanoxifen as and 
vhen it becaae available during the course of the studies. In turn ve 
require that each group nov ensures that this increased risk is 
appropriately reflected in the trial protocol and consent fora and 
that adequate aeasures are in place for aonitoring the occurrence of 
endonetrial changes during the trial. 

Although the NSABP PI protocol and consent forn does address this 
issue, I vould ask that you, upon reviev of this docuaent, re-exaaine 
if both the protocol or consent fora needs to be amended. 

I nust remind you that the continued supply of trial aedication to the 
NCI depends upon the adoption of these aeasures. 



98 



Page 2 

Drs. Fisher and Ford 



As I nentioned on the phone, ve believe that it vould be appropriate 
to convene a workshop Involving not only representatives of the 3 
prevention trials but also a nunber of independent experts in the 
fields of epideaiology, breast cancer and gynecology, so that ve can 
discuss together the impact of the information available on 
risk-benefit assessment for vomen participating in these trials. Ve 
vill be progressing this over the next 2-3 months. In the meantime, 
obviously ve felt it very important that you see this information. 



B«st regards. 
Sincerely, 




Paul V. Plourde, M.D. 
Director, Endocrine Resemrch 



PVP:mc 



99 



nrrERNAL kemorardun 

ZENECA Pharmaceuticals Group 
ntm 30-Jun-1993 lOsSOaa EOT Dni^ Ragulatory Affalra 

Ttltphon* (302)886-2127 
TOt Sc« Bclov 

nUMi Anthoay F. Rogart 

SUBJICTi FDA CONTACT-NOLVAOBZ 

AT 10 1 20 TODAY I 00NTACT8D MR. PAUL ZIMMERMAN (FOA-CSO) TO ADVISB HM OF THE 
FOLLOVINGt 

1-OBS ISSUE I SEHBCA IS PREPARING A DOCUMENT VHICH PROVIDES ALTERNATITB 
LANGOAGB AND JUSTIFICATION. THIS DOCUMENT VILL BE AVAILABLE LATE NEXT 
VEER AND I WILL CALL BIN VHEN I RBdEVE IT. EB AGREED. 

2-ENDO CBANGBSi BASED ON CURRENT RBVIBV OF DATA ZBHBCA VILL MODIFY THE PIB TO 
UPDATE TBB LANGUAGE DBSOUBING ENDOMETRIAL CHANGES. BE ASK FOR THE LANGUAGE 
AND I READ TO HIM THE PROPOSED BID STRESSING TBAT IT WAS DRAFT AMD COULD 
CHANGE BUT THE MESSAGE VOULO NOT. I ALSO STATED TBAT DR. PLOURDB VILL CALL Oft. 
nSHBR UTBR IN TBB DAT TO ADXVSB HIM OF TBB LABEL CHANGE. BE ASKED IF 
ANYTHHIG VILL CHANGE REGARDING THE PREVBHnOM TRIAL. I SAID TBAT OUR POSITION 
IS THAT TBB RISK/BENEFIT RATIO FOR BREAST CANCER TRBATMBNT DOBS NOT C8AM3 AND 
THAT THE FRBVBNnON TRIAL S0OUID CONTINUE BUT TBAT VAS CK. FISHER'S DECISION. 
VB BELIEVE TBAT TBB ISSUE OF BRDOMBTRIAL FIMDIN68 OPPOSITB NOLVADU TRBATMBNT 
SAVE BEEN PROFERLT ASSESSED IN THE DBLIBBRATIOHS REGARDING TBB PRBVBNTIOH 
TRIAL AND TBAT TBB LABEL CBAN6B SISBNGBTEMS TBB LANGUACB IN THE PIB. BE ASKED 
IF EBNBa VAS GOING TO DO ANYTBING FUKTBER REGARDING TBI8 ISSUE . I AP VI8BD 
TBAT IN THE NEAR FUTURE VB VILL OOHVENE A PANEL OF EXPERTS TO REVIEV THE DATA 
AND TBAT I VILL ADVISB PDA VEEN T BI8 VIL L TAKE PUCE. BE SAID TBAT HE VILL 
ADVISB nt. JUSTICE (FDA-MEDICAL RBVIBVBR) OF THE LABEL CHANGE. I TOLD MR. I 
TBAT IF TDk NBBOBD ANYMHtl SVC TO CONTACT ME. BE AGBB8D. 

TONT 

Distrlkotlaoi 

TOi Jack 6. IHineaa 
JU O'Sbaa 
Donald R. Vard 
Orlando Caasar 
Jaffragr A. Separ 
Stavan B. Laipart 
Rabaeea D. Cintran, M.O. 
* Paul V. Plourda 
Karan S. Linaa 
Vllllan C. Loena 
Vllllan J. Kannady 
Banish Canaroa 
Jaany Bolnaa 
Ronald L. Rrall 



SCIENTIFIC MISCONDUCT IN BREAST 
CANCER RESEARCH 



WEDNESDAY, JUNE 15, 1994 

House of Representatives, 
Committee on Energy and Commerce, 
Subcommittee on Oversight and Investigations, 

Washington, DC. 

The subcommittee met, pursuant to notice, at 10 a.m., in room 
2123, Rayburn House Office Building, Hon. John D. Dingell (chair- 
man) presiding. 

Mr. Dingell. The subcommittee will come to order. 

In April of this year, the subcommittee held a hearing on a series 
of issues involving the University of Pittsburgh's and Dr. Bernard 
Fisher's mismanagement of and the National Cancer Institute's 
failure to oversee, some of the Nation's most important clinical 
trials involving the treatment and prevention of breast cancer. 
Some of the issues discussed at that hearing included the falsifica- 
tion and fabrication of data at St. Luc Hospital in Montreal, Can- 
ada; the failure of the University of Pittsburgh, Dr. Fisher, and 
NCI to deal with the fraud in a timely manner; the failure of Dr. 
Fisher, NCI, and the Office of Research Integrity (ORI) to inform 
the women of America of the fraud for nearly 3 years; Dr. Fisher's 
failure to publish a reanalysis of the data, which was required to 
be done because of the need to exclude the fraudulent St. Luc data 
excluded; the failure of Dr. Fisher to inform NCI, Zeneca, which is 
the pharmaceutical manufacturer of tamoxifen, and the American 
women in a timely manner about the endometrial cancer deaths 
due to tamoxifen; and the failure of Dr. Fisher and his colleagues 
to inform NCI about audit findings revealing significant data irreg- 
ularities at a second Montreal hospital, St. Mary's. 

These multiple and serious failings came to a head in March of 
this year, when the media broke the news and the American public 
finally learned what NCI, the University of Pittsburgh, and Dr. 
Fisher should have told them long ago. And yet, these revelations 
are only a part of a long story. 

First, new problems have been disclosed at several additional 
sites. In March 1994, NCI found reports of audits conducted in No- 
vember 1992 at Tulane and LSU. These National Surgical Adju- 
vant Breast and Bowel Project reports, that is, NSABP reports, 
prepared one year following the audits, revealed serious and chron- 
ic problems of missing and misrepresented data. In fact, the major- 
ity of the data at these two sites could not be located. 

(101) 



102 

After two further NCI audits at these two sites, significant ques- 
tions remained about the completeness and the accuracy of the 
data. And the audit process continues. 

In two other instances, the subcommittee identified audit re- 
ports, one dating from 1990 and one from 1992, both of which re- 
vealed very significant discrepancies and irregularities. At Memo- 
rial Cancer Research Foundation, the NSABP auditor reported, 
among numerous other items, that, "A serious problem has been 
identified with this institution with respect to the accuracy of the 
data reported to the NSABP at the time of randomization." Fur- 
ther, the NSABP audit report cited a "serious problem" at the site 
because, and I quote again, "IRB, Institutional Review Board, will 
not approve or reapprove the NSABP protocols." 

At South Nassau Communities Hospital, two of eight patients re- 
ported as eligible were found by the audit to be clearly ineligible 
due to their medical history. The auditor recommended, and I 
quote, "additional randomization be suspended from this institution 
until such time as the principal investigator and his associates de- 
velop procedures which ensure eligibility, treatment, and follow-up 
of all patients entered into NSABP protocols." 

Curiously, neither of these problem-ridden audits received any 
effective follow-up. Indeed, in the case of the Memorial Cancer Re- 
search Foundation, the original audit report was found in a file 
drawer at Pittsburgh. The report had never been sent to NCI or to 
the research site. 

The NCI recently completed an audit at the Foundation, and as 
a result, NCI has referred the site for formal investigations of sci- 
entific misconduct and violation of Department of Health and 
Human Services regulations for the protection of human research 
subjects. 

As for South Nassau, contrary to NSABP's claims to NCI, it was 
not suspended. No effort has been made to determine how wide- 
spread the observed discrepancies were. Neither was any effort 
made to determine if patients had been harmed by their entry into 
protocols for which they were not eligible. Further reviews of South 
Nassau are now pending. 

Moreover, the subcommittee's review of hundreds of audit reports 
revealed that, in many cases, NSABP was years behind in perform- 
ing audits and in writing up and forwarding audit reports after the 
audits were completed. More significantly, the follow-up to identi- 
fied audit deficiencies was all but nonexistent. 

NSABP itself has recently begun to review its own past audits 
and reported several dozen sites with major discrepancies that 
have not been resolved yet. Dr. Fisher and his colleagues have 
claimed this was due to a lack of resources for an adequate audit 
staff. They claimed that they asked for additional resources from 
NCI, but were turned down. NCI says that this is not so. 

The revelations prior to and at the April hearing, as well as sub- 
sequently, have triggered a number of further developments: 

Dr. Fisher has been removed as head of NSABP; 

Some of Dr. Fisher's top colleagues at the University of Pitts- 
burgh, under the cloak of anonymity, started what appears to have 
been an ill-advised letter- writing campaign. They were demanding 
an investigation of the NCI Director, Dr. Samuel Broder, and the 



103 

reinstatement of Dr. Fisher, even in the face of mounting evidence 
of significant management failures; 

At the recommendation of NCI, the Office of Research Integrity 
has directed Pittsburgh to conduct an inquiry to determine if a sci- 
entific misconduct investigation is warranted involving the actions 
of Dr. Fisher and others at NSABP; 

And the grant for the overall administration of NSABP will be 
recompeted next year. 

The senior vice chancellor for health sciences at the University 
of Pittsburgh told the subcommittee staff he believes there is a cul- 
ture in the scientific community of inappropriate deference to the 
superstars of science. He believes that this culture has resulted in 
numerous institutions failing to carry out their lawful responsibil- 
ities in overseeing the work and actions of prominent scientists. 
The University of Pittsburgh now says that this culture of def- 
erence was a significant factor in its failure to oversee the work of 
Dr. Fisher. 

This same problem has become apparent to a number of other in- 
stitutions and has been found in a number of other questions ex- 
amined by this subcommittee. We hope that the scientific commu- 
nity as a whole will come to recognize that this is a serious prob- 
lem. 

Second, for years there have been questions about the role of 
pharmaceutical company funds in clinical research. Here, very sub- 
stantial sums of money, with little or no accounting for the expend- 
iture of that money, passed from Zeneca to NSABP, and to the Uni- 
versity of Pittsburgh. It will be noted that a precise accounting has 
not been obtained from any of the parties. 

In turn, the manufacturer gained substantial benefits via even- 
tual food and drug approvals, publicity, and similar events. The es- 
timated figures show that Zeneca gave the University of Pittsburgh 
some $600,000 for a chair for Dr. Fisher. Zeneca also gave NSABP 
about $600,000 for their semiannual meetings. Zeneca also sup- 
plied to NSABP about $300,000 for other purposes. This is a total 
of over $1.5 million, plus the substantial cost of supplying the 
tamoxifen drug free of charge for the trials. 

Essentially, at the same time the audit program was floundering, 
Zeneca was providing huge sums of money, not for audit resources, 
but for lavish parties and receptions at NSABP's semiannual meet- 
ings. These meetings were held in splendid places around the Unit- 
ed States and Canada. For example, the meetings were in places 
which imposed severe hardship on attendees such as the Doral 
Country Club; Banff Springs Hotel in the Canadian Rockies; Cha- 
teau Frontenac in Quebec City; the Fairmont Hotel in San Fran- 
cisco; Hilton Head, S.C; Vancouver, British Columbia; Palm 
Springs, Calif.; DisneyWorld and Bal Harbour, Fl. 

Some of the receptions included, among other things, strolling 
troubadours, wine-tasting parties, fire dances, live steel bands, and 
fine food, champagne, and dance. At Bal Harbour, they touted an 
elegant evening at the seashore with premium hard liquor, beer, 
and wine. A number of these receptions cost up to $80,000 apiece. 
Curiously, in 1991, the entire NSABP audit function was carried 
out on a budget of a little more than $80,000. 



104 

Third, today we will hear about the failure to report in a timely 
manner the deaths from endometrial cancer of patients receiving 
tamoxifen in Dr. Fisher's studies. We will also examine the matter 
of informed consent for the tamoxifen prevention study. While this 
study was in progress and after Dr. Fisher and his colleagues 
learned of at least two deaths in the study, deaths ultimately prov- 
en to be due to endometrial cancer, a sentence was added to the 
informed consent that said, and I quote now, "No deaths from 
endometrial cancer have been reported." This sentence was not re- 
moved from the consent form until 1994, at which time Dr. Fisher 
had learned of the death of at least four persons related to 
endometrial cancer. 

We look forward to hearing Dr. Fisher's account and explanation 
of these matters. 

In concluding the opening statement, the Chair wants to remind 
all of us of what is most important in this affair. This matter is 
part about failed responsibility. It is part about failures of scientific 
integrity. But above all, the question that should concern us most 
is the welfare of the thousands of women who, at some considerable 
risk to themselves, committed their lives and health to these stud- 
ies. 

The fact that thousands of women were involved, the fact that 
statistical power was so great that, overall, the conclusions may not 
be changed, does not mitigate the sacrifice of the women who were 
entered into studies for which they were not eligible or for which 
they did not give consent or on which they were not properly in- 
formed. It does not compensate for the risks these women faced, 
only to have their data discarded due to abject sloppiness, gross 
negligence or even fraud. 

As we proceed today, the Chair reminds us all that it behooves 
us to be mindful of, and to honor, these women. 

The Chair recognizes the gentleman from Ohio. 

Mr. Brown. Thank you, Mr. Chairman. I want to thank you 
again for holding this follow-up hearing on scientific fraud in feder- 
ally funded breast cancer research. It is obviously an important 
issue that warrants this subcommittee's and this Congress's imme- 
diate attention. 

What is it we are really talking about today? We are talking 
about the health and well-being of our wives, our mothers, our 
daughters. Women all over America are doubting their choices and 
questioning their health decisions. We owe it to them to give them 
answers on which they can depend. 

Unfortunately, we are currently stuck in a holding pattern. With 
each new audit, it appears there is new information about bad or 
missing data, about inappropriate enrollment of women in the 
study, about inaccurate reporting and follow-up. In a study of this 
magnitude, at some point you are likely to encounter a few simple 
errors in the collection of information. However, the extent to 
which there have been errors and to which there have been false 
data uncovered is alarming; and I believe it calls for complete in- 
vestigation. 

In addition, it is clear that NSABP has been quite lax in its own 
auditing to ensure clean and useful information. 



105 

Mr. Chairman, I don't believe that any of us wants to bring down 
the scientific establishment or damage the reputation of NSABP. 
At the same time, we owe it to all women to ensure that the infor- 
mation they use in consultation with their physicians is truthful, 
is reliable, and is up to date. 

Thank you, Mr. Chairman. 

Mr. DiNGELL. The time of the gentleman has expired. 

The Chair recognizes now the distinguished gentleman from Col- 
orado, the Ranking Minority Member of the committee, for such 
opening statement as he chooses. 

Mr. SCHAEFER. Thank you, Mr. Chairman. As the subcommittee 
and the chairman know, I began our April hearing on this matter 
by stating that we must seek to reassure the American people 
about the soundness of American science. Two months have passed, 
and although we have made some progress towards this goal, we 
still have a lot to accomplish. 

Thanks to your leadership, Mr. Chairman, this subcommittee 
acted promptly in exploring the circumstances surrounding re- 
search activities of the NSABP. It did so in a calm, professional 
manner because, while it is vital that we uncover fraud and mis- 
management, it is also essential that we calm the fears of Amer- 
ican women who relied on the studies in this question. 

We are relieved to learn that the basic results of the study con- 
taining falsified data have since been verified by subsequent analy- 
sis. I am troubled, however, to find that the problems outlined at 
our April hearing were more the rule rather than the exception. 

The subcommittee staff has done a terrific job in uncovering 
these problems. It has learned that extensive reporting problems 
across the United States and Canada existed. The NSABP did send 
competent and able auditors into the field, it seems. However, it 
failed to act upon the information received from its audit teams, 
neither generating appropriate reports nor notifying the institu- 
tions of potential problems. 

This situation was unacceptable in April and it certainly is 
today. The Food and Drug Administration and the NCI have al- 
lowed breast cancer prevention trials to resume; and while I agree 
this research is definitely important, given the track record of 
NSABP, I have very, very serious concerns. 

Now, Mr. Chairman, I look forward to hearing what actions have 
been taken by the University of Pittsburgh since our last hearing, 
as well as learning what the National Cancer Institute has done 
in light of the testimony this subcommittee received. 

This hearing is not about an institution or research program, it 
is about millions of American women who have a right to be able 
to rely on scientific and medical information given to them by their 
government; and that is why these hearings are so important, Mr. 
Chairman. I certainly applaud you for continuing this effort. 

Yield back. 

Mr. DiNGELL. The Chair thanks the gentleman. 

The Chair wants to observe that the staffs of the Minority and 
the subcommittee have worked very closely on this matter, as they 
always do, and I want to express my appreciation to my friend 
from Colorado and to his very able staff for their assistance in this 
matter. 



106 

The gentlewoman from Pennsylvania. 

Ms. Margolies-Mezvinsky. Thank you, Mr. Chairman, thank 
you for having these hearings and continuing to do so. 

Mr. Chairman, I would like to thank you for all that you have 
done with regard to the National Surgical Adjuvant Breast Project. 
The NSABP is an important, federally funded program which was 
designed to study treatments for breast cancer and bowel cancer, 
as we know. I am pleased that we will be hearing from several im- 
portant witnesses today regarding their role in the recently discov- 
ered auditing problems with the program. 

The question on all women's minds is whether the conclusions of 
these studies should be trusted. I want to make sure that we can 
assure women across this country that underlying recommenda- 
tions for treatment of breast cancer have not changed. Instead, this 
hearing will focus on problems with breast cancer programs and 
will ensure that in the future such mistakes will not be made 
again. 

I think that we have some fundamental questions that must be 
answered, and I hope that they will be answered here today; that 
is, how did it happen, how can we prevent this from happening 
again, how can we in general be more vigilant, and how can we an- 
swer questions for folks like Jill Sigal, who is in the audience here 
today, when she asked, can women who choose lumpectomy over 
mastectomy be comfortable with their decisions? 

I look forward to hearing from the panel. Thank you once again, 
Mr. Chairman, for holding these hearings. 

[The opening statement of Ms. Margolies-Mezvinsky follows:] 

Statement of Hon. Marjorie Margolies-Mezvinsky 

Mr. Chairman, I would like to thank you for holding this second hearing on the 
National Surgical Adjuvant Breast Project (NSABP). The NSABP is an important 
Federal funding program which was designed to study treatments for breast cancer 
and bowel cancer. I am pleased that we will hear from severeil important witnesses 
today regarding their role in the recently discovered auditing problems with the pro- 
gram. 

The question on all women's minds is whether the conclusions of these studies 
should be trusted. I want to assure women across America that the underlying rec- 
ommendations for treatment of breast cancer have not changed. Instead, this hear- 
ing will focus on problems with breast cancer program and will ensure that, in the 
future, such mistakes will not be made again. 

I am pleased that there will four panels testifying today, including representatives 
from Zeneca Pharmaceuticals, Dr. Bernard Fisher, the former director of NSABP, 
representatives from the University of Pittsburgh, and Dr. Samuel Broder, Director 
of the National Cancer Institute. 

In the past few months, we have heard press accounts of several serious auditing 
problems with certain sites in these studies. These auditing problems seem to be 
pervasive in certain sites of the study. I am disturbed that the information about 
faulty data was not shared with the National Cancer Institute in a more timely 
manner. I understand that new procedures have been put into place to ensure that 
this detections of faulty data will be given to both the National Cancer Institute, 
the NSABP, and officials with Zeneca. All parties must be fully informed of any 
problems to ensure that women can trust the result of these trials. 

In addition, I am pleased that the National Cancer Institute is conducting its own 
auditing procedures to determine if further problems there are in the NSABP. These 
confirmations of previous auditing are necessary to ensure the soundness of the con- 
clusions of clinical trials. 

Finally, I am pleased that the clinical preventive trials on tamoxifen have been 
restarted. Many women are interested in the results of these preventive trials. I be- 
lieve that the Congress must continue to be vigilant to ensure that proper auditing 
standards are being applied to all clinical sites. It is my understanding that a new 



107 

auditing system will be used in these preventive trials to ensure that all women in 
these trials have been given a consent form, that their clinical information has been 
confirmed, and that proper precautions are being taken to prevent further abuses 
in this vital research project. 

I look forward to learning more about this situation and will continue to be vigi- 
lant in my efforts to assert the facts regarding this research project. The women of 
America deserve to know the facts, and I will make sure that those facts are avail- 
able to them. 

Mr. DiNGELL. The Chair thanks the gentlewoman for her opening 
statement. 

The Chair announces that the first panel will be composed of Dr. 
John Patterson, M.D., international medical director, Zeneca Phar- 
maceuticals Group, accompanied by Dr. Paul Plourde, senior direc- 
tor, clinical and medical affairs, Zeneca Pharmaceuticals Group, 
and Alan Milbauer, vice president, external affairs, Zeneca Phar- 
maceuticals Group. 

Gentlemen, welcome to the subcommittee. 

Gentlemen, as you are no doubt aware, it is the practice of this 
subcommittee that all witnesses testify under oath. Do any of you 
object to so doing? 

The Chair advises that, given the fact that you are testifying 
under oath, it is your right to be advised by counsel. Do any of you 
desire to be advised by counsel during your appearance here? 

Very well, then gentlemen for your information, copies of the 
rules of the committee, the subcommittee and the House of Rep- 
resentatives are there in the red and the blue booklets before you 
at the table. 

Gentlemen, if you have no objection, then, to testifying under 
oath, would you please each rise and raise your right hand. 

[Witnesses sworn.] 

Mr. DiNGELL. Gentlemen, you may each consider yourself under 
oath. We will recognize you starting with Dr. Patterson; then. Dr. 
Plourde and Mr. Milbauer, you will be recognized for such state- 
ments as you choose to give. Gentlemen, proceed. 

TESTIMONY OF JOHN PATTERSON, INTERNATIONAL MEDICAL 
DIRECTOR, ZENECA PHARMACEUTICALS GROUP, ACCOM- 
PANIED BY PAUL PLOURDE, SENIOR DIRECTOR, CLINICAL 
AND MEDICAL AFFAIRS, AND ALAN MILBAUER, VICE PRESI- 
DENT, EXTERNAL AFFAIRS 

Mr. Patterson. Thank you, Mr. Chairman. Mr. Chairman and 
members of the subcommittee, I am Dr. John Patterson, inter- 
national medical director of Zeneca" Pharmaceuticals, working at 
our company's headquarters in the United Kingdom. Joining me 
today are Dr. Paul Plourde, senior director of clinical and medical 
affairs, and Alan Milbauer, vice president, external affairs, at 
Zeneca Pharmaceuticals Group, both based in Wilmington, Dela- 
ware. Zeneca is a bioscience-based business, formerly a part of Im- 
perial Chemical Industries, or ICI. 

Our company is proud of its history of dedication to scientific re- 
search and patient support in the field of women's health. Zeneca 
is the discoverer, developer and manufacturer of Nolvadex, the 
world's leading drug for the treatment of breast cancer and the 
world's most prescribed anticancer medicine. Nolvadex is our trade 
name for tamoxifen. Since 1973, millions of patients have taken 



108 

tamoxifen as an important component of their breast cancer treat- 
ment. 

In recent months, there has been considerable press regarding 
tainted data in breast cancer trials conducted by NSABP. It is un- 
derstandable that patients are wondering if the medical basis for 
choosing their treatment is correct. They have also heard that the 
risk of developing uterine cancer as a side effect of treatment with 
tamoxifen is greater than previously thought, and they wonder how 
that risk applies to them. Additionally, they have witnessed a heat- 
ed public debate over the merits of the use of tamoxifen to prevent 
breast cancer in healthy women who are at high risk of developing 
this deadly disease. 

We are concerned that public confusion over these issues could 
lead patients with breast cancer to be afraid of tamoxifen, a medi- 
cation with the demonstrated ability to delay recurrence of their 
disease and prolong their lives. 

Mr. Chairman and members of the subcommittee, let me be very 
clear: For treatment of breast cancer, tamoxifen is one of the most 
studied drugs in the world. Its safety and efficacy for this purpose 
is very well established. With over 6 million patient-years of expe- 
rience in more than 20 years of clinical use around the world, 
tamoxifen continues to be a key weapon in the battle against 
breast cancer. 

In patients with early-stage breast cancer, tamoxifen reduces the 
risk of recurrence of the disease after 10 years by up to Va, and it 
reduces the odds of death by 20 percent. Further, clinical trials of 
patients with early breast cancer have also demonstrated a marked 
reduction in new cancers affecting the other breast in women treat- 
ed with tamoxifen. 

In summary, the proven benefits of tamoxifen in the treatment 
of breast cancer in men and women outweigh its risks. It is my 
hope that these hearings will underscore this indisputable fact. 

As an anticancer agent, tamoxifen is a powerful medicine, and in 
recent years, an increased risk of developing uterine cancer has 
been reported in women who are taking it. Zeneca has closely mon- 
itored the incidence of uterine cancers, both from clinical trials and 
spontaneous reports. We have promptly reported the data to the 
FDA and have also presented it to independent experts for evalua- 
tion. 

Starting in 1989, Zeneca has amended its U.S. label on four dif- 
ferent occasions as new information about the risk of uterine can- 
cer was received. We have recent data collected from 14 studies in- 
volving nearly 17,000 women worldwide. They indicate that the in- 
creased risk of uterine cancer associated with tamoxifen is approxi- 
mately 20 times smaller than the more deadly risk that breast can- 
cer will recur or spread if patients do not take tamoxifen. 

Much has been made in the press about reports of uterine cancer 
deaths in the large B-14 trial involving tamoxifen in breast cancer 
patients being conducted by NSABP. To obtain accurate, long-term 
information, the women who participate in this trial are followed 
until they die. Thus, reports of deaths are an expected component 
of cancer clinical trials; and it is a sad fact, Mr. Chairman, that 
even with a potentially curable tumor such as uterine cancer, some 
women contracting this cancer will die of it. 



109 

We have been informed that one of the subjects of today's hear- 
ing will be the timeliness of reports about four uterine cancer 
deaths that occurred in the B-14 trial. The first of those deaths oc- 
curred on June 25, 1991. Zeneca was informed of that death on 
February 10, 1992, in a routine NSABP report. The cause of death 
was not clearly stated, and several potential causes were listed 
which included uterine cancer. 

We reported this case to the Food and Drug Administration on 
April 1, 1992. NSABP reported the remaining three uterine cancer 
deaths to Zeneca on December 13, 1993, and we reported these to 
FDA on January 5, 1994. According to the subcommittee staff, the 
NSABP apparently has claimed that it notified Zeneca of one of 
these deaths on February 1, 1993. 

In fact, NSABP provided Zeneca with a large data set, including 
data that reflected that death, but no cause of death was specified 
in that data. It was not until December 1993 that Zeneca learned 
from NSABP that this death was, in fact, due to uterine cancer. In- 
deed, NSABP's own August 1993 report to its membership failed to 
identify any uterine cancer death in any patient. 

The B-14 uterine cancer deaths have brought into question the 
wisdom of continuing tamoxifen in the prevention trial. I want to 
make it clear to this committee that Zeneca is not sponsoring or 
endorsing the tamoxifen breast cancer prevention trial. This trial 
is sponsored by the National Cancer Institute and is being con- 
ducted by NSABP. A brief history of the genesis of the prevention 
trial may be helpful to the committee. 

In 1985, Zeneca was strongly encouraged by a significant number 
of scientists in the medical community to sponsor a tamoxifen 
breast cancer prevention trial. We responded by organizing a con- 
ference involving leading experts in medical oncology, animal toxi- 
cology, and epidemiology. The consensus was that additional data 
were required before an informed decision could be made on the ap- 
propriateness of the prevention trial. As additional data became 
available, Zeneca provided it both to regulatory authorities and the 
scientific community. 

In 1990, members of the academic community again approached 
Zeneca about a possible prevention trial. The company reiterated 
its concerns. However, following many discussions, the National 
Cancer Institute, the FDA, and medical experts decided that the 
potential risks and benefits put together — that the benefits out- 
weighed the risks and that a study of tamoxifen in prevention 
should be conducted. 

Zeneca, then the sole supplier of tamoxifen in the United States, 
subsequently agreed, at NCI's request, to provide tamoxifen tablets 
and matching placebo for the study. The tablets were supplied free 
of charge on the understanding that participants were to be prop- 
erly monitored and fully informed of the potential risks. NSABP 
also agreed to notify Zeneca of any significant new findings in the 
study. 

We have been informed that another subject of today s hearing 
will be Zeneca's financial relationships with the University of Pitts- 
burgh and the NSABP. Let me assure you, Mr. Chairman, that our 
relationship with the University and NSABP was at all times prop- 
er and consistent with applicable legal, ethical, and moral stand- 



110 

ards. Any insinuation to the contrary is completely false. I have 
outlined the full extent of these relationships in my written state- 
ment to the subcommittee. 

In summary, pajrments were made by Zeneca to NSABP for three 
categories of work; namely, data processing and management, sup- 
port for meetings, activities, and honoraria and travel costs to Drs. 
Fisher and Redmond in respective appearances at FDA and sci- 
entific meetings. In addition, Zeneca agreed to partly fund, at the 
request of the University of Pittsburgh, a chair in the Department 
of Surgery at Pittsburgh. Zeneca's conduct in this respect was en- 
tirely consistent with both the spirit and letter of FDA policy and 
the codes of conduct adopted by the American Medical Association 
and the Pharmaceutical Research Manufacturers Association, not 
to mention our own internal corporate guidelines, as they applied 
then and as they apply now. 

We believe that our financial assistance to the University of 
Pittsburgh and NSABP contributed to breast cancer research in the 
United States and fostered a better understanding of breast cancer 
treatment around the world. We do not believe that this financial 
support caused NSABP to be less diligent in reporting safety infor- 
mation regarding tamoxifen. In fact, much of our financial support 
was given specifically to obtain extra safety reports to submit to 
the Food and Drug Administration and to facilitate the exchange 
of information amongst breast cancer researchers. 

In conclusion, Mr. Chairman, we at Zeneca are committed to the 
development and support of safe medicines backed by good science 
and proper scientific practices. We believe that tamoxifen is a safe 
and effective medicine vital to the treatment of breast cancer. We 
deplore fraud and poor scientific practices in clinical research. We 
accordingly are pleased to assist the subcommittee in this inves- 
tigation. 

Thank you for the opportunity to speak. My colleagues and I 
would now be happy to answer your questions. 

[The prepared statement of Dr. Patterson follows:] 



Ill 



STATEMENT OF JOHN PATTERSON 

Mr. Chairman and Members of the Subcommittee, I am Dr. John Patterson, 
International Medical Director of Zeneca Pharmaceuticals, working at our company's 
headquarters in the United Kingdom. Joining me today are Dr. Paul Plourde, Senior 
Director of Clinical and Medical Affairs, and Alan Milbauer, Vice President External 
Affairs at Zeneca Pharmaceuticals Group, both based in Wilmington, Delaware. Zeneca is a 
bioscience based business, formerly a part of Imperial Chemical Industries, or ICI. 

Our company is proud of its history of dedication to scientific research and patient 
support in the field of women's health. Zeneca is the discoverer, developer and 
manufacturer of Nolvadex, the world's leading drug for the treatment of breast cancer and 
the world's most prescribed anticancer medicine. Nolvadex is our trade name for tamoxifen. 
Since 1973, millions of patients have taken tamoxifoi as an important component of thdr 
breast cancer treatment 

In recent months, there has been considerable press regarding tainted data in breast 
cancer clinical trials conducted by NSABP. It is understandable that patients are wondering 
if the medical basis for choosing their treatment is correct. They have also heard that the 
risk of developing uterine cancer as a side effect of treatment with tamoxifen is greater than 
was previously thought, and they wonder how this risk applies to them. Additionally, they 
have witnessed a heated public debate over the merits of the use of tamoxifen to prevent 
breast cancer in healthy women who are at high risk of developing this deadly disease. We 
are concerned that public confusion over these issues could lead patients with breast cancer to 



112 



2 



be afraid of tamoxifen, a medication with the demonstrated ability to delay recurrence of 
their disease and prolong their lives. 

Mr. Chairman and Members of the Subcommittee, let me be very clean For 
treatment of breast canc^, tamoxifen is one of the most studied drugs in the world. Its 
safety and efficacy for this purpose are very well established. With over six million patient- 
years of experience in more than 20 years of clinical use around the world, tamoxifen 
continues to be a key wes^n in the battle against breast cancer. In patients with early stage 
breast cancer, tamoxifen reduces the risk of recurrence of the disease after 10 years by up to 
one third, and reduces the odds of death by 20%. Further, clinical trials of patients with 
early breast cancer have also demonstrated a marked reduction in new cancers affecting the 
other breast in women treated with tamoxifen. Data also suggest that tamoxifen may 
decrease cardiovascular disease and stabilize postmenopausal bone loss. In summary, the 
proven benefits of tamoxifen in the treatment of breast cancer in men and women outweigh 
its risks. It is my hope that these hearings will underscore this indisputable £act 

As an anticancer agent, tamoxifen is a powerful medicine and, in recent years, an 
increased risk of developing uterine cancer has been reported in women who are taking it 
Zeneca has closely monitored the incidence of uterine cancers both from clinical trials and 
spontaneous reports. We have promptly reported that data to the FDA and have also 
presented it to independent experts for evaluation. Starting in 1989, Zeneca has amended its 
label on four different occasions as new information about the risk of uterine cancer was 
received. We have recent data collected from 14 studies involving nearly 17,000 women. 
They indicate that the increased risk of uterine cancer associated with tamoxifen is 



113 



3 
q}pFO]uinately 20 times smaller than the more deadly risk that breast cancer will recur or 

spread if patients do not take tamoxifen. 

Much has been made in the press about tepoits of uterine cancer deaths in the large 
B-14 trial involving tamoxifen in breast cancer patients that is being conducted by NSABP. 
To obtain accurate, long-term information, the women who participate in this trial are 
followed until they die. Thus, rqwrts of deaths are an ejq>ected component of cancer clinical 
trials, and it is a sad fact that, even with a potentially curable tumor such as uterine cancer, 
some women contracting this cancer will die. 

We have been informed that one of the subjects of today's hearing will be the 
timeliness of rqxirts about four uterine cancer deaths that occurred in the B-14 trial. The 
first of those deaths occurred on June 25, 1991. Zeneca was informed of that death on 
Fd}ruary 10, 1992 in a routine NSABP rqx)rt. The cause of death was not clearly stated 
and several potential causes were listed including uterine cancer. We rqrarted this case to 
the FDA on April 1, 1992. NSABP rqwrted the remaining three uterine cancer deaths to 
Zeneca on Decemb» 13, 1993. We reported these deaths to tfie FDA on January 5, 1994. 
According to the Subcommittee staff, the NSABP apparently has claimed that it notified 
Zeneca of one of these deaths on February 1, 1993. In fact, NSABP provided Zeneca with a 
data set reflecting that death, but no cause of death was specified. It was not until £>ecember 
1993 that Zeneca learned from NSABP that this death was due to uterine cancer. Indeed, 
NSABP's own August 1993 report failed to identify any uterine cancer death in any patient 

The B-14 uterine cancer deaths have brought into question the wisdom of continuing 
the tamoxifen prevention trial. I want to make it clear that Zeneca is not sponsoring or 



114 



4 

endorsing the tamoxifen breast cancer prevention trial. This trial is sponsored by the 

National Cancer Institute (NCI) and is being conducted by NSABP. A brief history of the 
genesis of the prevoition trial may be helpful: 

In 1985, Zeneca was strongly encouraged by a significant number of scientists in the 
medical community to sponsor a tamoxifen breast cancer prevention trial. Zeneca responded 
by organizing a conference involving leading experts in medical oncology, animal toxicology 
and epidemiology. The consensus was that additional data were required before an informed 
decision could be made on the appropriateness of a prevention trial. 

As additional data became available, Zeneca provided it to regulatory authorities and 
the scientific commimity. In 1990, members of the academic community approached Zeneca 
again about a possible prevention trial. The company reiterated its concerns. Following 
many discussions, the NCI, the FDA and medical experts decided that the potential benefits 
of prevention outweighed the risks and that a study of tamoxifen in prevention should be 
conducted. 

Zeneca, then the sole supplier of tamoxifen in the United States, subsequently agreed, 
at NCI's request, to provide tamoxifen tablets and matching placebo for the study. The 
tablets were supplied free of charge on the understanding that participants were to be 
properly monitored and fiilly informed of the potential risks. NSABP agreed to notify 
Zeneca of any significant new findings. 

As a company, Zeneca has a major commitment to research not only on breast cancer 
but also on other conditions that affect women's health. We have participated and will 
continue to participate in all appropriate discussions of the scientific issues and facts that 



115 



5 
have a bearing on the prevention and treatment of breast cancer. 

We have been informed that another subject of today's hearing will be Zeneca's 
relatiooshq) with the University of Pittsburgh and the NSABP. Let me assure you that our 
relationship with the University and the NSABP was at all times proper and consistent with 
applicable legal, ethical and moral standards. Any insinuation to tiie contrary is completely 
false. I would like to outline for the Subcommittee the full extent of these relationships. 

Zeneca first developed a working relationship with NSABP in the mid-1970's, after 
tamoxifen had been approved for use in the U.K. but before it had been introduced in the 
United States. When NSABP approached us regarding the use of tamoxifen in trials, they 
were one of the leading breast cancer research groups in the world, with an unequalled 
rq>utation for conducting large scale cancer clinical trials. 

Over diese nearly twenty years, the NSABP has conducted a large number of trials on 
tamoxifen, all of which were developed, funded, s^roved and conducted under the auspices 
of the NCI. With the excq)tion of the prevaition trial, all previous NSABP tamoxifen 
studies have been undertaken with large groups of women who have been diagnosed and 
treated surgically for breast cancer. 

Starting in 1982, we contracted with the NSABP to obtain data from the trials they 
were conducting on tamoxifen. These data were used to support our applications to the FDA 
for new indications for tamoxifen and to comply with FDA safety and monitoring 
requirements. Since 1982, payments from Zeneca to NSABP for the cost associated with 
data management, analyses and reports on studies of tamoxifen have totaled approximately 
$200,000. An additional $73,000 was paid for work done on tamoxifen at the University of 



116 



6 

Kansas. Contracts such as these are routine in our industry. They are necessary to gain 

access to clinical research findings, data and analyses. 

From the available records, Zeneca paid Dr. Bernard Fisher and Dr. Carol Redmond, 
about $11,000 over neaiiy twenty years to participate in educational programs and to appear 
before the FDA in support of our drug applications and filings. These activities are 
consistent with the accq)ted practices in our industry. 

Over this same period of time, NSABP solicited funding from Zeneca and several 
other companies to help support NSABP meetings. According to available records, Zeneca's 
contributions towards these activities from 1976 through 1993 total approximately $585,000. 

The purpose of the NSABP meetings was and is to bring NSABP cooperative 
researchers and others in the cancer research community together to share the latest 
information concerning cancer research. NSABP meetings were intense working sessions 
which covered topics related to the day to day problems and situations inherent in the 
running of, or the recruitment for, large clinical trials. NSABP controlled the agenda for 
their meetings and Zeneca had no influence over the scientific program. Continuing 
education credit was often available for attendance at these meetings. 

Industry funding for dinners and receptions is a common and long accq)ted practice 
within this country's scientific community. Zeneca's conduct was entirely consistent with 
both the letter and spirit of FDA policy and the codes of conduct adopted by the American 
Medical Association (AMA) and the Pharmaceutical Research and Manufacturers Association 
(PhRMA), not to mention our own stringent internal corporate guidelines as they applied 
then and now. 



117 



7 
Finally, in 1988 at the request of the University of Pittsburgh, Zeneca agreed to 

participate in the funding of a research chair at the University. The purpose of this chair 

was to assist the University of Pittsburgh's Dqiartment of Surgery in their stated desire to 

attract and recruit graduate students and faculty members interested in pursuing researdi in 

oncology in honor of Dr. Fisher. Zeneca has donated $600,000 to help establish the chair. 

It is our understanding diat sufficient funds have y^ to be raised to complete the $1.S million 

endowment and the chair has not been created or filled. 

We believe that our financial assistance to the University of Pittsburgh and NSABP 
contributed to breast cancer research in the U.S. and fostered a better understanding of breast 
cancer treatment around the world. We do not believe that this financial support caused 
NSABP to be less diligent in rqwrting safety information regarding tamoxifen. In fact, 
much of our financial support was given specifically to obtain extra safety rq>orts to submit 
to the FDA and to facilitate the exchange of information among breast cancer researchers. 

In conclusion, we at Zeneca are committed to the development and support of safe 
medicines backed by good science and proper scientific practices. We believe tamoxifen is a 
safe and effective medicine vital to the treatment of breast cancer. We dq)Iorc fraud and 
poor scientific practices in clinical research. We accordingly are pleased to assist the 
Subcommittee in this investigation. 

My colleagues and I would be happy to answer any of your questions. 



118 

Mr. DiNGELL. Thank you, Dr. Patterson. 

Dr. Plourde. 

Mr. Plourde. I do not have a prepared opening statement, Mr. 
Chairman. 

Mr. DiNGELL. We would be glad to recognize you for any com- 
ments you might like to make at this time. 

Mr. Plourde. No, I think Dr. Patterson adequately reflected our 
comments. 

Mr. DiNGELL. Very well. 

Mr. Milbauer. 

Mr. Milbauer. Mr. Chairman, I don't have a prepared state- 
ment, either. I am just here to help answer questions that the com- 
mittee may have. 

Mr. DiNGELL. Very well, gentlemen. Thank you for your prepared 
statement. 

The Chair recognizes the distinguished gentleman from Colorado 
for questions. 

Mr. SCHAEFER. Thank you, Mr. Chairman. 

Dr. Patterson, before we begin, I would like to give you an oppor- 
tunity to allay the fears of some of the general public. Is tamoxifen 
effective in treating women with breast cancer? 

Mr. Patterson. There is a very simple answer to that. Yes, sir, 
it is effective both in the advanced disease — and data on that has 
been generated over many years in many countries in the world — 
it is effective in treating the early disease as a so-called adjuvant 
therapy. 

Those adjuvant therapy studies have been conducted also in 
many parts of the world. For instance, the NSABP data only rep- 
resents approximately 10 percent of the major studies that have 
been conducted worldwide on adjuvant therapy of this stripe. Even 
if you were to remove all of the NSABP studies from the world 
overview of adjuvant therapy, the conclusions on efficacy and safety 
would remain the same. 

Mr. SCHAEFER. How many studies have been conducted? Are we 
talking about a large number, small number? How many women 
have been involved? 

Mr. Patterson. There are many hundreds of trials of this study 
throughout the world. In adjuvant therapy, for instance, there are 
approximately 40 major studies worldwide; but in the advanced 
disease and in other treatments, many hundreds. 

Mr. SCHAEFER. Well, the news reports have been focusing not on 
the treatment trials but on the prevention trials; is that correct? 

Mr. Patterson. That appears to be the case, sir. 

Mr. Schaefer. That appears to be the case? 

Mr. Patterson. The safety of tamoxifen for prevention has been 
the subject of much media discussion in this country. 

Mr. Schaefer. I understand that — if I do understand it correctly, 
prevention trials involved giving tamoxifen to women with a high 
risk of getting breast cancer, but who, in fact, have not really con- 
tracted the disease; is this correct? 

Mr. Patterson. That is correct. They simply do not have the dis- 
ease. 

Mr. Schaefer. You indicated in your opening statement that 
your company was asked to participate in prevention trials, but you 



119 

were reluctant to do so. I know that you covered this somewhat 
briefly in your opening statement, but I would like to give you an 
opportunity to expand on your answer. 

Why was it that you were reluctant to participate in these pre- 
vention trials? 

Mr. Patterson. When these were first mooted in 1985, we had 
concerns on a number of fronts, shared also by a number of sci- 
entists in the field. The drug had been developed as a medicine for 
cancer. Most of the studies had been in women with the advanced 
disease and had not been long term. It is not possible to do, for in- 
stance, placebo-controlled studies in advanced cancer, so the abso- 
lute profile of this agent was not absolutely clear; and whilst the 
animal toxicology was adequate for use as an anticancer agent, it 
certainly wasn't adequate for use in patients without cancer. So 
over the period 1985 to 1990, studies were done by the company. 

Also, clinical studies were done, including studies by NSABP, 
where data on all of these subjects matured. By 1990, there were 
data on controlled studies on women treated for long periods of 
time with the drug, who had not got advanced disease, and there- 
fore the side-effect profile could be recognized more clearly. 

We had further animal toxicology which had included some very 
serious findings in the rat liver, which were a concern to us and 
to the outside world. And we also had a whole series, or a whole 
increased amount of safety information from large numbers of pa- 
tients; and my company, in reviewing this situation, was concerned 
that although women at high risk of developing breast cancer clear- 
ly have a major worry as to whether they are going to develop the 
disease, and anything that one can do to help them has to be con- 
sidered, that 50 percent of the women who develop breast cancer 
do not come from that group; and that even when a woman is at 
high risk of developing breast cancer, a relatively small number of 
those women will develop the cancer over a relatively short period 
of time. 

And that meant that for the majority of the women taking part 
in such studies, all they could expect to get from a medicine such 
as tamoxifen was the side effects of this agent. It was unlikely to 
be protecting them, although there were some suggestions and 
there are some suggestions that it may prevent postmenopausal 
bone loss and it may also prevent cardiovascular diseases that re- 
sult from the effects of cholesterol. 

So it was a very difficult, risk-balanced equation; and that is 
clearly why it is being debated so hard in public. 

Mr. SCHAEFER. But you were asked by the medical community, 
I think, to participate in these prevention trials; were you not? 

Mr. Patterson. We were. 

Mr. Schaefer. And you have given your answer. The NCI made 
this request as well, didn't they? 

Mr. Patterson. NCI did make this request, made a request for 
the material for the trials. 

Mr. Schaefer. Given your consent, you made it conditional, 
upon reviewing the consent form, that it was to be given to women 
to participate in the trial; is that correct? 

Mr. Patterson. That is correct. 



120 

Mr. SCHAEFER. Some of these questions lead up to what I am 
going to be asking of witnesses in the future here. Did you, in fact, 
review the consent forms? 

Mr. Patterson. We reviewed a model consent form in the fall of 
1991, extensively within the company. 

Mr. SCHAEFER. This was the form used in the trials, or was it 
changed without your knowledge and consent? 

Mr. Patterson. The consent form appears to have evolved con- 
siderably over a period of time subsequent to this. This was the 
consent that was initially submitted, as I understand it, to the 
FDA by NSABP; and there were changes that occurred subsequent 
to that, which we did not review. 

Mr. Schaefer. OK, so there were changes. 

So you were basically unaware of the affirmative statement con- 
cerning uterine deaths that was at some point included in this con- 
sent form? 

Mr. Patterson. We did not rereview the consent form and find 
that statement. 

Mr. Plourde. Congressman, if I can add a statement to that 

Mr. SCHAEFER. Please. 

Mr. Plourde. Although we were not aware that this change had 
been made early on, I believe we did become aware toward the 
middle of 1993 that this change had been made. However, at the 
time, we were told by the NSABP that indeed that was a correct 
and true statement. 

Mr. Schaefer. Did you register any objection to this? 

Mr. Plourde. We did not. 

Mr. Schaefer. Was it a true statement? 

Mr. Plourde. We were told it was a true statement at the time 
that we became aware that this was in the consent form, by the 
NSABP. 

Mr. Schaefer. But is it a true statement, to your knowledge, at 
this point? 

Mr. Plourde. No, it is not. 

Mr. Schaefer. Let me turn to the issue of deaths arising from 
the trials. 

You requested notification of any deaths; is that correct? 

Mr. Patterson. That is correct. 

Mr. Schaefer. You were notified of all the deaths? 

Mr. Patterson. We were notified of all the deaths, yes. 

Mr. Schaefer. Approximately how many death notifications do 
you routinely receive in the course of a year? 

Mr. Patterson. Over the course of a year, over 100 patients are 
dying across the NSABP studies involving tamoxifen. We receive 
an annual printout or an annual computer diskette which contains 
data on large numbers of patients. The latest one contained more 
than 300 patients. 

Mr. Schaefer. When the deaths are reported to you, is the cause 
of a particular death indicated? 

Mr. Patterson. Deaths are reported to us in a number of dif- 
ferent ways. 

Again, perhaps just for clarification, women are expected to die 
in these studies. They are suffering from breast cancer. Death is 
not always an unexpected event, so deaths were reported to us as 



121 

routine or deaths were reported to us as a serious or unusual event 
or one that occurred on treatment; and there is a difference. 

Mr. SCHAEFER. So was the cause reported to you? 

Mr. Patterson. Causality was not linked to death, although — I 
am not trying to be difficult in answering your question. The way 
that the data were reported to us was that other factors associated 
with the death were listed in a separate column, but causality spe- 
cifically was not reported. 

Mr. SCHAEFER. Mr. Chairman, I see my light is on, and I will 
pass. 

Mr. DiNGELL. The time of the gentleman has expired. 

The Chair recognizes now the gentleman from Ohio, Mr. Brown. 

Mr. Brown. Thank you, Mr. Chairman. I would like to follow up 
on some of Mr. Schaefer's questions about tamoxifen and some of 
the — both the prevention — particularly the prevention trials. 

Dr. Patterson, based on what you said, I would infer that it is 
very important for Zeneca to be kept informed in a timely way of 
information regarding side effects being generated in Dr. Fisher's 
breast cancer treatment trials, correct? 

Mr. Patterson. That is correct. 

Mr. Brown. How and when did your company convey to Dr. 
Fisher, other officials in NSABP, the importance of timely and ac- 
curate information regarding any side effects of tamoxifen? 

Mr. Patterson. Over the course of the more than 14 years in 
which we were associated with studies that they were conducting, 
my company on many occasions interacted with Dr. Fisher and 
NSABP on the issue of side effects. We discussed these about them, 
we wrote memos to them, we had an agreement with them, the 
most recent of which was in 1986, on exactly what side effects data 
would be reported to us and when. 

Mr. Brown. One of the key side effects that Zeneca was con- 
cerned about was endometrial cancers, correct? 

Mr. Patterson. Correct. 

Mr. Brown. When were you first notified that at least one uter- 
ine cancer death had occurred with the cause attributed to 
tamoxifen? 

Mr. Plourde. Mr. Congressman, maybe I can try to answer that. 

As part of our annual obligation to report adverse events to the 
FDA, we requested a number of items from the NSABP; and on a 
yearly basis we got the following information: 

We asked them to provide us with a listing of deaths or with- 
drawals occurring while on treatment in those various trials that 
we held. In addition, we asked for a computer diskette on second 
tumors, and those tumors are malignancies that occur after the di- 
agnosis of breast cancer. In addition, we also take information ob- 
tained from the NSABP annual report to its membership, which 
contains adverse effects, contains narratives on serious adverse re- 
actions, and it contains information on overall survival. 

In February of 1992, we received information on a death in a pa- 
tient who had the diagnosis of endometrial cancer. We requested 
further information from the NSABP in regards to this patient, and 
I reviewed that information and assessed the death to be attrib- 
utable to endometrial cancer. However, one must keep in mind that 



122 

the cause of death is often very difficult to establish and involves 
some medical judgment. 

I did have a discussion with the NSABP in regards to this death, 
and it was their belief that this patient had died from other causes, 
other than endometrial cancer. Hence, I reported to the FDA the 
information that I received, which included the possible relation- 
ship to endometrial cancer as well as to pulmonary embolus. 

As Dr. Patterson mentioned in his opening statement, most re- 
cently we were informed that we allegedly had information on a 
second endometrial cancer death in February of 1993, and that was 
included in the computer diskette. The diskette information not 
only includes the second tumor data that occurred in that particu- 
lar trial, but also the survival status of the patient at the time of 
the report. There is no causality mentioned in that particular re- 
port, and it was not possible at that time, neither is it possible now 
that — in reviewing that data to really make an association between 
endometrial cancer and the death of this particular patient. 

We were dependent on the NSABP to medically review this infor- 
mation and to provide us with this data, so the first time, given 
the scenario that I have just stated, that Zeneca appreciated the 
deaths had occurred related to endometrial cancer was actually in 
December of 1993. 

Mr. Brown. So you are saying that the December 1993 case was 
where you were notified by NSABP that tamoxifen was the cause 
of the uterine cancer specifically, precisely? 

Mr. Plourde. No, I think in December of 1993 we were informed 
that the patient's death had been attributable to the endometrial 
cancer in a patient taking tamoxifen. 

Mr. Brown. How were you informed of that? 

Mr. Plourde. I was called by a member of the NCI and informed 
that there had been a death reported at the NSABP meeting. 

Mr. Brown. NSABP did not call you? NSABP said it at a meet- 
ing. Dr. Fisher said it at a meeting, NCI then called you? You have 
not to this day heard it directly from NSABP? 

Mr. Patterson. We subsequently have had information from 
NSABP, in December, listing the patients who have died and giv- 
ing the identifiers. 

Mr. Brown. What triggered that? You called NSABP, they ac- 
knowledged that was the case, but NSABP never initiated the call 
to you about that death or subsequent deaths? 

Mr. Plourde. That is correct. 

Mr. Brown. Is that the way NSABP should be operated? 

Mr. Plourde. No. They should be providing that information to 
us as soon as it is known to them. 

Mr. Brown. Why didn't they call you? What is your view of why 
they didn't call you? 

Mr. Plourde. It is difficult to speculate. I suspect that their pri- 
orities in regards to informing us were different than ours, but I 
really couldn't say why they didn't call. 

Mr. Brown. It is not like they didn't notify somebody they didn't 
know. You had this long-term, dozen-plus years relationship with 
them; they know you, you know them; you would over time empha- 
size the importance of that two-way exchange of information; they 



123 

simply failed to notify a very important colleague, if you will, of 
theirs? 

Mr. Patterson. It is conceivable that they didn't recognize these 
as endometrial cancer deaths until December 1993, but you would 
have to ask NSABP. 

Mr. Brown. Once Zeneca learned of these deaths, what were 
your responsibilities and whom did you inform as your duty to the 
public? What was the next step once you found this sort of indi- 
rectly through NCI? 

Mr. Patterson. Our step is to inform the FDA in a timely fash- 
ion, and as necessary, or if necessary, to alter our package insert 
with the statement then that is read by all doctors using the drug. 

Mr. Brown. Did the information about the deaths necessitate 
any changes in the consent form for either the prevention studies 
or the treatment studies? 

Mr. Patterson. The consent form for the prevention studies has 
been changed to remove the statement that there have been no 
deaths. 

Mr. Dingell. The time of the gentleman has expired. 

The gentlewoman from Pennsylvania. 

Ms. Margolies-Mezvinsky. Clarify something for me. From 
what we understand, the NSABP presented the information in Oc- 
tober at a conference. Tell me where the time lapse is. 

Mr. Plourde. There was an annual NSABP meeting in October 
of 1993 where the endometrial cancer data was presented. I was 
in attendance of that meeting. However, a lot of information was 
being conveyed; there were two projectors going on simultaneously. 

And I was taking copious notes, and I did not note at that par- 
ticular time that this particular death had apparently been men- 
tioned in the presentation; and this was duly noted by a member 
of the NCI. 

Ms. Margolies-Mezvinsky. So in other words, what happened 
was that, at some point, the information had not been thoroughly 
gone through, and it fell through the cracks? 

Mr. Patterson. Can I answer that, Congresswoman? 

It would be very unusual for any group to report a death to us 
or to any other person by means of standing up in a public meeting 
or trial — a meeting such as that one. 

Furthermore, unless that patient's trial number were given, we 
couldn't identify that patient and deal with it appropriately, so we 
would expect to be notified in writing or by a telephone call di- 
rectly, not to just happen to be sitting in a meeting where some- 
thing like that is presented. 

Ms. Margolies-Mezvinsky. Do you know why you were not noti- 
fied? 

Mr. Patterson. No, I don't. 

Ms. Margolies-Mezvinsky. Dr. Patterson, once Zeneca had 
learned of these deaths, what was your responsibility and whom 
did you inform — did you turn to with regard to these deaths? 

Mr. Patterson. The first responsibility, as a pharmaceutical 
company, is to notify the Food and Drug Administration; and we 
did that in early January of 1994. 

The second responsibility is to inform doctors or patients associ- 
ated with use of the drug of that occurrence if it is germane to 



124 

their treatment. In patients who are treated with tamoxifen for 
breast cancer, we had recognized since 1989 that endometrial can- 
cer, uterine cancer was occurring; and it is recognized that 15 to 
20 percent of women who develop endometrial cancer will die of 
that disease, as I said in my opening statement, so some deaths 
were not unexpected. 

The issue really relates here to how use of the drug for breast 
cancer can be translated across into a prevention situation where 
the odds of death are significantly different. 

Ms. Margolies-Mezvinsky. What about labeling? Did it have 
any effect on labeling? 

Mr. Patterson. The labeling was changed in April of this year 
to include the statement — on the fact that there has been a death. 

Ms. Margolies-Mezvinsky. Did the information about the 
deaths necessitate any changes to the consent forms for either the 
prevention or the treatment studies? 

Mr. Patterson. It clearly had an effect on the prevention con- 
sent form, and I believe that has been changed. In the studies of 
more advanced breast cancer, statements have been in the consent 
for a number of years stating that uterine cancer was a potential 
risk of taking tamoxifen; and it has usually been included in the 
potentially fatal toxicity section of the protocol. 

I don't believe there were any statements in most of those con- 
sent forms — although Dr. Plourde may want to add to this — about 
the fact that there had been no deaths, and some deaths would re- 
grettably be expected. 

Ms. Margolies-Mezvinsky. In determining the then existing 
consent form needed to be changed, were you surprised at all to 
learn that the contents of the consent form were what they were? 

Mr. Patterson. Yes. 

Ms. Margolies-Mezvinsky. Could you elaborate, please? 

Mr. Patterson. Although we had reviewed a consent form in the 
fall of 1991, that had not contained a statement that there had 
been no deaths. That statement had been introduced subsequently, 
and we had not reviewed subsequent consents. Those consents had 
been sent to the company, but with covering letters detailing the 
changes to both protocol and consent; and there was no — our atten- 
tion was not drawn to that fact, so it was not — it was a surprise 
to us that there was a problem in this area. 

Ms. Margolies-Mezvinsky. Do you know who added the state- 
ment? 

Mr. Patterson. I do not know. 

Ms. Margolies-Mezvinsky. What about the existing consent 
form for the prevention study, what about it stood out to you? 

Mr. Patterson. I don't think I understand your question. 

Ms. Margolies-Mezvinsky. What about the existing consent 
form for the prevention study stood out? 

Mr. Patterson. The one that exists today? 

Ms. Margolies-Mezvinsky. Yes. 

No, the one that existed then. 

Mr. Patterson. At the time, when the company reviewed it, the 
company felt that from its knowledge of the toxicity of tamoxifen, 
it was quite a good consent form; but there were a number of 
things that were in our package insert that were not fully covered 



125 

in that consent, and Dr. Plourde wrote to NSABP informing of that 
at the time that we reviewed it. 

Ms. Margolies-Mezvinsky. The consent form used at that time 
had a statement in it that no endometrial cancer deaths had been 
reported due to the use of tamoxifen, correct? 

Mr. Patterson. At the time the study was initiated in June 
1992, my understanding is, yes, that statement was in there. 

Ms. Margolies-Mezvinsky. Was that statement in the original 
version of the consent form for the prevention study? 

Mr. Patterson. I am sorry, I didn't hear you. 

Ms. Margolies-Mezvinsky. Was that statement in the original 
version of the consent form for the prevention study? 

Mr. Patterson. The one that we reviewed, it was not in it. 

Ms. Margolies-Mezvinsky. Thank you very much. 

Mr. DiNGELL. The time of the gentlewoman has expired. 

The Chair now recognizes the gentleman from California, Mr. 
Moorhead. 

Mr. Moorhead. Thank you, Mr. Chairman. 

Did approximately the same number of women take the placebo 
that did the tamoxifen? 

Mr. Patterson. Which study are you referring to, Congressman? 

Mr. Moorhead. The study that you have been referring to. 

Mr. Patterson. The Study B-14? 

Mr. Moorhead. Yes. 

Mr. Patterson. Yes, that is correct. 

Mr. Moorhead. Have any of the women who took the placebo 
come down with uterine cancer? 

Mr. Patterson. There have not been any deaths from uterine 
cancer in those patients, to our knowledge. 

Mr. Moorhead. What percent of the women that took the 
tamoxifen have come down with uterine cancer? How many have 
taken it and how many came down with the illness? 

Mr. Patterson. It is about 25 out of 1,400 are the numbers that 
come to my mind, but I don't have those numbers in my head. Con- 
gressman. 

Mr. Moorhead. Twenty-five out of 1,400. Of those women that 
took the placebo, how many eventually got breast cancer? They 
were people 

Mr. Patterson. Had a recurrence of their disease? 

Mr. Moorhead. Yes. 

Mr. Patterson. Again, sir, I don't have those numbers. 

Mr. Plourde. Congressman, maybe I can provide some informa- 
tion on that. 

Two patients in the B-14 trial- 



Mr. Patterson. Did you ask of breast cancer or 

Mr. Moorhead. What I am trying to see are the benefits versus 
the risks; and if the tamoxifen patients were saved from cancer, 
from breast cancer in substantial numbers, I want to see whether 
the risk was worthwhile. Do you see the direction I am going? 

Mr. Patterson. Yes, I do, indeed. 

Mr. Plourde. Unfortunately, all of the adjuvant breast cancer 
studies, when they were initiated, were not designed to provide us 
with information on endometrial cancer. When this was determined 
in follow-up, this was investigated quite carefully. Unfortunately, 



126 

due to the various biases of the trial, because they weren't set up 
in proper form, it is difficult to assess as to the incidence of breast 
cancer in a placebo group. In fact, the two patients that were ran- 
domized to placebo in the B-14 trial did receive tamoxifen subse- 
quently and had breast cancer. 

Mr. Patterson. Can I try and answer your question for you? 

There are less women who developed recurrent breast cancer in 
the placebo group than in the treatment group. The absolute num- 
bers, I don't know off my head, but about Vs less developed recur- 
rent breast cancer in the treatment group than in the placebo 
group. That approximated — if you look then at deaths, where there 
was a 20 percent reduction compared with the — for the tamoxifen 
group compared to the placebo, there have been 4 deaths from 
endometrial cancer. We are talking about a 20-fold difference be- 
tween those two groups, so we are talking about, I think, 80 deaths 
versus 4 deaths. 

Mr. MOORHEAD. I would think those figures in that information 
would be something that was very important to give to women that 
might be considering using the drugs so that they could weigh the 
risks one way or the other. 

Mr. Patterson. I would agree with you. 

Mr. MoORHEAD. Dr. Patterson, on several occasions. Dr. Fisher 
requested funding from Zeneca. Did he ever ask for funding for im- 
proving his administrative staff? 

Mr. Patterson. Not to my knowledge. 

Mr. MoORHEAD. But did he ask for funding for receptions at the 
NSABP meetings? 

Mr. Patterson. Yes, he did. 

Mr. MOORHEAD. Did he get that? 

Mr. Patterson. Yes, he did. 

Mr. MOORHEAD. If he had asked for funding for administrative 
staff, do you think he would have gotten it from you? 

Mr. Patterson. It is in our interests and everybody's interests 
to ensure that the data from these studies are as good as they can 
be; and I am sure we would have responded positively to that kind 
of request. 

Mr. MOORHEAD. You have had a total of 25 deaths now from 
uterine cancer? 

Mr. Patterson. A total of 25 patients have developed uterine 
cancer. There are only four deaths. 

Mr. MOORHEAD. I see. Can you tell whether it came from the — 
how many of these women would have gotten uterine cancer 
whether they had taken tamoxifen or not? 

Mr. Patterson. That is a difficult question. The incidence of 
uterine cancer as a background in the population is one way that 
you can look at that, but the control group is the other; and there 
are actually no uterine cancers in the control group which, in itself, 
is surprising. You would have expected in that number of women 
of that age group over that period of time to have had some of 
those occur, but that has not been the case. 

Mr. Moorhead. That is why I was interested in knowing how 
many of those who had taken the placebo had eventually come 
down with the uterine cancer. I think that would be something that 



127 

would help you understand the pluses and minuses of taking the 
drug. 

Mr. Patterson. We have calculated from our worldwide 
databases how the incidence of endometrial cancer on patients tak- 
ing tamoxifen varies from the background incidence in the popu- 
lation at large. Patients with breast cancer do have an increased 
incidence of uterine cancer anyway, but it looks to us as if the in- 
crease associated with tamoxifen is somewhere between two- and 
fivefold over the background incidence, taking all of the studies we 
are aware of worldwide together. 

Mr. MOORHEAD. I guess the last question — and you probably 
can't answer it — but because of the benefits of tamoxifen, obviously 
not liking the dangers that come to the few who take it, is it pos- 
sible to find any other drug or any change in tamoxifen that might 
eliminate that risk? 

Mr. Patterson. Yes, it is possible to try and find an antiestrogen 
that would be called a pure antiestrogen, that is, it had no effects 
on the uterus of the kind that we are describing; and that has been 
the subject of research in my company and a number of others for 
many years. It is possible that such agents will be developed or 
even that a partial agonist antiestrogen, of which tamoxifen is one, 
that didn't have those particular positive effects on the uterus 
might also be possible to be developed; and there may even be one 
or two in very early-stage research at the moment. 

Mr. Moorhead. You are dealing with a subject that is of great 
concern to every husband in the world, I am sure, because of want- 
ing to protect their wives, and when things like this come up you 
wonder why it gets members of the committee excited. It is because 
they are concerned for their loved ones. 

mr. Patterson. It touches us all. Congressman. 

Mr. Moorhead. Thank you, Mr. Chairman. 

Mr. Brown [presiding]. Thank you, Mr. Moorhead. 

Dr. Patterson, returning to the 1993 time period that both I and 
Mr. Schaefer earlier had asked about, the subcommittee obtained 
records out of NSABP depicting a slide presentation dated in Au- 
gust of 1993, not just the October, and we talked about the October 
seminar, but a slide presentation August, 1993. And that slide 
presentation there are at least two deaths attributable to — two 
deaths that were attributed from that cancer, from endometrial 
cancer, in the B-14 study with patients taking tamoxifen. 

It should be in front of you. One is numbered 006252, where a 
patient died, and the other one was number 006254. Were you 
aware of this information in August, 1993? 

Mr. Patterson. I am not aware that any member of the com- 
pany received that slide presentation or that we were informed of 
it by the NSABP. 

Mr. Brown. Why did it take 4 months? We know that was pre- 
pared in August. We are not saying that you knew about it in Au- 
gust. We know it was prepared in August. Why did it take that — 
take 4 months from the time these slides were prepared to inform 
you, as the manufacturer, of these deaths? 

Mr. Patterson. That is a question you must ask NSABP. I don't 
know how to answer that. 



128 

Mr. Brown. Isn't it true that by the summer of 1993, when the 
information was being prepared for these slides, if they were actu- 
ally prepared by August that if the information was being prepared 
in the summer that Zeneca was pressing NSABP regarding any in- 
cidence of endometrial cancers associated with tamoxifen? Were 
you doing that? 

Mr. Patterson. Yes. We had been conducting worldwide review 
of this particular subject. We had pressed NSABP very hard for 
their up-to-date information, and we were convening a panel of ex- 
perts to help us to review the situation on endometrial cancer. 

Mr. Brown. So you don't know why this information, then, 
wasn't provided to you as you were holding these conversations 
with them? 

Mr. Patterson. No, we don't nor why it wasn't recognized in 
their consent form for prevention. 

Mr. Brown. In your testimony you indicated that Zeneca had 
provided upwards of a million and a half dollars in direct support 
to the University of Pittsburgh to NSABP for a variety of projects. 
Was that money solicited or unsolicited? Did Zeneca volunteer it or 
did Pittsburgh and NSABP request it? 

Mr. Patterson. The money, which was paid over a period of 
some 20 years, divides into several different categories, some of 
which were requested, some of which were volunteered. 

For moneys for data handling and analyses that we required of 
them to allow us to put our FDA applications in, that was money 
that we offered to them in return for the work that they had to do 
over and above their normal study work to achieve those analyses. 

Mr. Brown. So did you feel obliged to provide them those funds 
in some manner? 

Mr. Patterson. That would be absolutely normal practice for us 
to do that, yes. 

Mr. Brown. What kind of benefits did you enjoy by providing 
these funds? 

Mr. Patterson. The funds for the data handling allowed us to 
provide information to the FDA on the beneficial effects of 
tamoxifen in Studies B-9 and B-14 in particular, and also to pro- 
vide, as far as we could, a risk-benefit analysis by providing side 
effects and adverse reaction information on those studies and on all 
other tamoxifen studies conducted by NSABP. 

Mr. Brown. Did you know that some of that money, to the tune 
of hundreds of thousands of dollars, were going to some of the par- 
ties and other entertainment that Chairman Dingell mentioned? 

Mr. Patterson. That was by separate request, so the data for — 
the data handling was separately requested for the data 

Mr. Brown. Separately requested by Dr. Fisher or by whom? 

Mr. Patterson. By the NSABP, yes. They didn't request the 
money for the data handling. We requested the information for 
which we offered money for them to provide us with that scientific 
information. The money for the receptions at the NSABP annual 
meetings or biannual meetings was requested of us by NSABP for 
that specific purpose. 

Mr. Brown. You say receptions. Chairman Dingell made it look 
a little more elaborate than receptions. But you knew that money 
was to be used for that purpose? 



129 

Mr. MiLBAUER. Yes, Congressman. If I could answer that, as you 
know or as perhaps you don't know, NSABP would hold biannual 
and then it became annual meetings for the purpose of bringing to- 
gether all of their cooperative researchers to be able to discuss over 
a period of 3 or 4 days how trials ought to be run, what new proto- 
cols ought to be considered, discuss the analyses of information 
that would go on. And these were heavy scientific sessions that 
would take place over these 3 or 4 days. 

On one evening they would hold — they would want to hold a so- 
cial event. And we were asked, and it began almost 18 years ago, 
if we would sponsor such an event. These were receptions. These 
were receptions that were not sit-down dinners. They did provide 
food, hors d'oeuvres, if you will, and cocktails, but these were — ^this 
was an integral component of the opportunity for these researchers 
to get together in an informal session, exchange information. 

We didn't control and had no involvement with the agenda for 
such meetings, and these are all within the guidelines that are 
published by the American Medical Association with respect to 
making sure that the substantial majority of the time devoted to 
these scientific meetings were devoted to the sessions and that the 
expenses were modest and that they did facilitate exchange. And 
within the spirit of that, that is what we were sponsoring over 
these many years. 

Mr. Brown. Modest. Two things come to mind when you say they 
were modest, and you continue to use the word reception. I have 
not been to a lot of receptions that cost in the rough ballpark of 
$50,000 to $80,000, which several of these cost. And, second, these 
weren't held in Pittsburgh. They were held in Hilton Head. They 
were held in the Canadian Rockies. 

I am not disputing your right as a private corporation to do that. 
I am just questioning your labeling of some of these as educational, 
as routine, as simple receptions. 

Mr. MiLBAUER. Well, Congressman, let me address that for you. 
It is common practice between the pharmaceutical industry and 
scientific institutions and medical institutions to sponsor and help 
in the conduct of those meetings, but I think it is important to dis- 
tinguish. We did not choose the venue for the meetings. The venue, 
whether it was in Bal Harbour or in San Francisco, that was cho- 
sen by the NSABP. We had no role 

Mr. Brown. But they also knew they had a sugar daddy to help 
them defray the expenses. 

Mr. MiLBAUER. The only expenses we believe to our knowledge 
that was defrayed in terms of administrative expenses, because it 
is a nonprofit organization, was the reception that evening. Our 
funds, to the best of our knowledge, were not utilized in the accom- 
modations for the attendees nor for the transportation to those 
meetings. 

And, in fact, I think it also was important when you look at a 
$50,000 or an $80,000 reception is that the numbers of attendees 
at these meetings grew over the years. That is, there were only 
about 6 or so receptions of some almost 30 that were in that, but 
the number of attendees approached 1,000 or even more than that 
on some occasions. 

Mr. Brown. I would yield to you, Mr. Schaefer. 



130 

Mr. SCHAEFER. Two quick questions. Were any Federal moneys 
involved in the sponsoring of these receptions? 

Mr. MiLBAUER. To the best of my knowledge, there were no Fed- 
eral moneys provided for the reception. In fact 

Mr. SCHAEFER. How about transportation, lodging, et cetera? 

Mr. MiLBAUER. I don't know. It would be whatever NSABP would 
fund. They may have gotten that from NCI, but we were actually 
advised by the NSABP that they could not get funding from the 
Federal Government for the reception. And it was on that basis — 
they would ask other companies to help sponsor the meetings, and 
this was the way 

Mr. SCHAEFER. I am talking about the reception alone. I mean 
the transportation down, the lodging, everything else. 

Mr. MiLBAUER. To the best of my knowledge, there was no Fed- 
eral funding for the reception. 

Mr. SCHAEFER. For the reception only. All right. And you paid 
100 percent of the reception? 

Mr. MiLBAUER. Yes. We would pay those to the hotel or we would 
pay those to a meeting planner. We did not pay those moneys to 
the NSABP or anyone else. It was just for those arrangements. 

Mr. SCHAEFER. All right. Well, we will ask questions about trans- 
portation and lodging later. 

Mr. Brown. I would add that the transportation to these places 
other than Pittsburgh was apparently borne by the Federal Gov- 
ernment. 

One further question about this, and then I have a brief set of 
questions. You recently dropped your contribution to NSABP from 
the $50, $60, $70, $80,000 range down to $10,000. Yet you make 
these statements that they were routine, and they were simply re- 
ception. Why did you drop that amount? 

Mr. MiLBAUER. I believe that the meeting of $10,000 was a sec- 
ond meeting in 1993, and I think it was just — I don't recall wheth- 
er the request was for — in fact, I don't know what reception took 
place at that meeting, but we just funded it to $10,000. 

The prior time we actually just provided them with a grant be- 
cause during that time it was unclear as to whether the kinds of 
sponsorship of these meetings would be within the guidelines be- 
cause the guidelines were evolving. 

Mr. Brown. So, number one, what was the amount of the grant? 
Second, are you going back up to the $50,000 to $80,000? Is the 
$10,000 just a temporary drop just prior to this hearing and then 
back to the 50,000, 80,000? 

Mr. MiLBAUER. The $10,000 was provided in 1993. 
There is actually a meeting that is taking place at the present 
time. We were requested to provide the typical sponsorship. 

We received a request earlier this year. At the time, we deferred 
responding to that, and I think, despite the fact that we think 
there is no impropriety at all, in fact we have reviewed the guide- 
lines. We keep looking at them to determine that they are within 
the guidelines. We believe they are, but given the publicity and 
given the fact that there was going to be this hearing today, we 
had declined in providing support to this meeting, and in fact 
NSABP's response to that is, given the situation, they would not 
want to put themselves into a position of receiving moneys for that. 



131 

But I would say affirmatively on the basis of what the guidelines 
and how they operate, we see nothing wrong with what we have 
done in the past. 

Mr. Brown. To change the subject, is your parent company ICI? 
Is there still affiliation with ICI? 

Mr. Patterson. No, we are now a separate company. Zeneca is 
a totally separate share holding company. 

Mr. Brown. When was ICI — when was the disaffiliation, if you 

will? 

Mr. Patterson. In 1991. I am sorry, 1993. My apologies. 

Mr. Brown. ICI is to my understanding, if you can help me with 
this, the manufacturer of DDT and a distributor of DDT? 

Mr. Patterson. I don't think that I can answer for what ICI 
does. I am not an employee of ICI. I am a physician working in a 
pharmaceutical company. 

Mr. Brown. I understand that. I am only asking if, in fact, that 
is the case. If these companies were affiliated, I would think that 
you would know whether or not ICI manufactured DDT. 

Mr. MiLBAUER. It would be a speculation on our part, but I be- 
lieve that they do not manufacture DDT, but I can't say with any 
certainty whether ever in its history that they have. 

Mr. Brown. OK. I was just curious because the information that 
I have been given is that ICI does, in fact, manufacture DDT and 
that I find it curious that companies that are affiliated — one is 
doing such good research on breast cancer as you have done, and 
if you are not — if you have no knowledge of whether it manufac- 
tures DDT and whether DDT— if there is any scientific proof of 
whether DDT actually increases the risk of breast cancer, we will 
drop it at that. 

Mr. MiLBAUER. I think probably the best thing — and I would 
agree with Dr. Patterson — is for the committee to receive some in- 
formation from ICI with respect to whether they do or they don't. 

Mr. Brown. Other questions? 

Thank you, gentlemen 

I am sorry, Mr. Schaefer, the gentleman from Colorado. 

Mr. Schaefer. Two quick questions. Then I want to jump back 
to this 1992 death. After you made the determination that the 
death resulted from uterine cancer, you did notify the FDA; is that 
correct? 

Mr. Plourde. That is correct. 

Mr. Schaefer. Did you notify Dr. Fisher? 

Mr. Plourde. I don't recall if I spoke to Dr. Fisher or to his as- 
sistant, Dr. Wickerham. 

We do know — and we have records that we obtained further in- 
formation in this case from Dr. Wickerham. Upon review of that in- 
formation I did discuss the case with Dr. Wickerham, who felt that 
this patient had actually died of a pulmonary embolus rather than 
endometrial cancer. 

However, given the problems in accurately interpreting the cause 
of death, I reported to the FDA that it was possible that this 
woman had died from endometrial cancer, contributed by pul- 
monary embolus. 

Mr. Schaefer. Let me get this straight. You said that you don't 
recall whether or not you notified Dr. Fisher. 



132 

Mr. Plourde. The NSABP was notified, but Dr. Fisher himself, 
I can't say for certain that I notified him personally of that. 

Mr. SCHAEFER. You can't remember that? 

Mr. Plourde. I can't recall. I know I have had discussions with 
the NSABP, Dr. Wickerham, in this matter. 

Mr. SCHAEFER. Should the NSABP have been able to determine 
the cause of this patient's death? 

Mr. Patterson. The NSABP had access to the full case record 
forms, access to the hospital where the patient died and was treat- 
ed. We only had access to a one-paragraph summary. 

Mr. SCHAEFER. I understand that. I am asking a question, 
though. The NSABP — could they have determined the cause of the 
death? 

Mr. Patterson. They can determine the cause of death. They 
could have as per their judgment, but cause of death is not an easy 
judgment medically. 

Mr. SCHAEFER. Should they have done this? 

Mr. Patterson. With a retrospectroscope, they clearly have de- 
termined that to be the cause of death, but at the time they did 
not. 

Mr. SCHAEFER. They did not. Thank you. 

Mr. Brown. Thank you, Mr. Schaefer. 

Gentlemen, thank you for your testimony. Appreciate your being 
here. 

The Chair will now call up the second panel: Dr. J. Dennis O'Con- 
nor, chancellor. University of Pittsburgh; Dr. Ronald Herberman, 
interim chairman. National Surgical Adjuvant Breast and Bowel 
Project; and Dr. Thomas Detre, senior vice chancellor for health 
sciences. 

Dr. O'Connor, Dr. Detre, Dr. Herberman, welcome. In front of 
you are the rules of the subcommittee, the committee, the full com- 
mittee, and the House of Representatives. Feel free to consult these 
throughout. 

You are allowed certainly to have legal counsel here, and it is the 
tradition of this committee to testify under oath, if you would raise 
your right hand. 

[Witnesses sworn.] 

Mr. Brown. Please be seated. 

Dr. O'Connor, would you begin with your opening testimony, 
please? 

TESTIMONY OF J. DENNIS O'CONNOR, CHANCELLOR, UNIVER- 
SITY OF PITTSBURGH, ACCOMPANIED BY THOMAS DETRE, 
SENIOR VICE CHANCELLOR OF HEALTH SCIENCES, AND 
RONALD B. HERBERMAN, INTERIM CHAIRMAN, NATIONAL 
SURGICAL ADJUVANT BREAST AND BOWEL PROJECT 

Mr. O'Connor. Thank you, sir. 

Mr. Chairman and members of the subcommittee, I am Dennis 
O'Connor, chancellor of the University of Pittsburgh, and I sin- 
cerely thank you for accommodating my request to appear here 
today. I asked to come because I want to state in person that, as 
the senior executive officer of the University of Pittsburgh, I accept 
responsibility for the past administrative deficiencies of the Na- 
tional Surgical Adjuvant Breast and Bowel Project, which is 



133 

headquartered at our university. I also accept and welcome the re- 
sponsibility to make things right. 

I have stated these commitments to Dr. Broder, the Director of 
the National Cancer Institute, as well. 

I deeply regret and apologize for any anxiety induced in cancer 
patients and the general public as a result of this matter. I am par- 
ticularly sensitive to this unfortunate outcome because my wife is 
an oncology nurse who specializes in helping women with breast 
cancer to cope with the stress of their condition. 

As my colleagues, Drs. Detre and Herberman, will describe to 
you, a large team consisting of many of the University's most able 
people, as well as outside experts, have been working diligently to 
make the changes that we believe are warranted to justify restored 
public confidence in us and to advance this important and historic 
research. 

Mr. Chairman, as you know, there has been no shortage of public 
comment on this matter, with several hundred news accounts pub- 
lished and broadcast already. The media commentary is, on bal- 
ance, healthy, for it heightens public awareness of the issues and 
the awareness of the University faculty and administrators. But I 
do want to identifv two points that some of the news accounts may 
have obscured rather than clarified. 

While it is gratifying and reassuring that, as far as is known, 
NSABP's research findings continue to be entirely sound, charges 
of deficient administration are not adequately answered by merely 
asserting that the research findings remain valid. Falsified data is 
odious under any condition. 

Second, while credit for the intellectual achievements of a faculty 
member goes to that faculty member, the university is ultirnately 
responsible for the administration of research. All honor that is due 
Dr. Bernard Fisher and his colleagues for their achievement in this 
path-breaking breast cancer research is their honor. We fully recog- 
nize, however, that the responsibility for competent research ad- 
ministration ultimately rests with us at the University. 

To learn and to teach from observation and experience is the 
central mission of a University and the essence of science. What 
are some of the lessons we can learn from this experience? What 
can we now teach based on this experience? These large and dif- 
ficult questions are being considered intensively. I am not yet 
ready to offer a complete list of possible answers, but I am ready 
to suggest several. 

First, we must establish better mechanisms to ensure that even 
our most experienced, tested and accomplished faculty are respon- 
sive to administrative imperatives as well as to their pressing and 
sometimes compelling intellectual agendas. 

We must find better ways to promote an environment in which 
faculty understand that the University's proper oversight in areas 
of administrative compliance neither conflicts with academic free- 
dom nor collides with institutional respect for faculty members' 
judgment. 

Those of us who have an administrative role in the universities 
must constantly remind ourselves that while loyalty to colleagues 
is important, depending on the facts of the case, accountability to 
the public supersedes that relationship. 



134 

We must be very sensitive to the importance of accurately dis- 
closing in a timely manner emerging research developments that 
carry potential to affect the public adversely, even where there may 
be some uncertainty as to the facts. 

We must elicit from research sponsors clear statements of their 
needs and expectations. When we receive such statements from re- 
search sponsors, we must follow the statements punctiliously. If we 
disagree with the statements, we must take the initiative to 
achieve a meeting of the minds. If the statements are unclear, we 
must get them clarified. Ambiguity in matters of administrative re- 
sponsibility can undermine relations with research sponsors and 
indeed disserve the public. 

The role of this subcommittee and its staff in this matter re- 
quires special note, and I will speak candidly. Mr. Chairman, as 
you yourself have stated, this subcommittee has a reputation for 
sharp teeth. Few relish being investigated by this subcommittee. 
During the last 2 months we have had a fairly grueling set of inter- 
actions, document productions, interviews and information de- 
mands from the subcommittee staff. This work has been painstak- 
ing and not particularly pleasurable. 

However, I also wish to acknowledge that the subcommittee and 
its staff, once the facts were collected, dissected and analyzed, did 
not shrink from working very constructively with the National 
Cancer Institute and the University to get the NSABP research 
back on track, to preserve the best of NSABP, and hopefully enable 
us to improve the rest. 

If, in the end, the approach taken by this subcommittee had been 
negative and destructive rather than positive and constructive, 
breast cancer patients, present and future, would have been sub- 
jected to further suffering. Instead, they are being helped because 
of the advancement of a healthier partnership between NSABP and 
NCI that prevailed before. This subcommittee has been instrumen- 
tal in that. 

We are learning that sharp teeth can be painful and embarrass- 
ing but, unfortunately, can also be necessary from time to time. 

We will not claim today, Mr. Chairman, to know with mathe- 
matical precision exactly the right mix of breathing room and over- 
sight that is required at a University to produce great science, to 
generate medical knowledge, to save lives and alleviate suffering. 
Toward that ideal balance we strive imperfectly. Be assured, how- 
ever, that the knowledge that we are and will always be imperfect 
does not discourage us. This episode motivates us to work harder 
to achieve these goals, which we share with the Federal Govern- 
ment and the American people. 

Thank you, sir. 

Mr. DiNGELL. Thank you. Dr. O'Connor. 

Dr. Detre. 

TESTIMONY OF THOMAS DETRE 

Mr. Detre. Mr. Chairman and members of the subcommittee, I 
am Dr. Thomas Detre, senior vice chancellor for Health Sciences. 

Mr. DiNGELL. Doctor, the Chair apologizes. We have a miserable 
public address system, as you will fmd. 

Mr. Detre. My apologies. Can you hear me now, Mr. Chairman? 



135 

I am here in response to your invitation to comment on the 
NSABP matter and to answer your questions. 

Mr. Chairman, after even more than 40 years in America, I still 
speak with an accent. When I came to the United States several 
years after World War II, physicians like myself who had accents 
were desperately needed because of the shortage of skilled sci- 
entists in America's University hospitals and laboratories. But now 
I am happy to report people like me are rather obsolete because 
today there are so many exceptionally able American scientists. 
Today, most of the time, American senior faculty train foreign stu- 
dents, as well as American students, not vice versa. That is the 
amazing progress of American science in our lifetimes. 

During the same period, as you know, along with spectacular ad- 
vances in biomedical research in the United States, the science of 
treatment evaluation also progressed rapidly. The multi-center 
clinical trial, of which NSABP is a renowned example, is univer- 
sally regarded as a premier approach for testing medical 
hypotheses. 

Dr. Bernard Fisher has spent his lifetime developing such trials 
to gauge the efficacy of breast cancer treatments. These trials nota- 
bly involved his historic discovery that lumpectomy followed by ra- 
diation is fully as effective as the more devastating and disfiguring 
radical mastectomy which, before Dr. Fisher's discovery, was the 
treatment of choice for many forms of breast cancer. 

This landmark finding, I might add, has since been confirmed 
independently by at least five large-scale clinical trials in the Unit- 
ed States and Europe. 

One of the advantages of a multi-center clinical trial over the 
usual study conducted in a single or several academic medical cen- 
ters is that the multi-center trial can test a very large number of 
patients in diverse geographical, socioeconomical and medical set- 
tings, making the results more readily subject to generalizations 
about the population at large. Biostatisticians also note the tend- 
ency of such large studies to be valid by reason of the studies' sta- 
tistical power. 

A drawback of the broad scale multi-center clinical trial, how- 
ever, is the difficulty in controlling the quality and uniformity of 
the data coming from numerous investigators in numerous centers, 
some of whom are not highly expert in conducting such research. 
This difficulty may explain some of the current issues relating to 
NSABP. I say explain, I do not say excuse. 

We still do not know nor does the National Cancer Institute, 
with complete and total thoroughness, exactly what fraction of the 
NSABP data were unreliable. We are still double checking, as is 
NCI. But based on everything we know to date — and it is exten- 
sive — and despite the wide-ranging questions about certain NSABP 
administrative practices, that no knowledgeable person has chal- 
lenged the continuing validity of the scientific conclusions. 

I report this conclusion to the subcommittee not to minimize con- 
cerns about our responsibility to administer by the highest stand- 
ards but to reassure breast cancer patients and their families, as 
have the responsible government agencies, that public health and 
safety are not compromised. The soundness of the scientific conclu- 
sions is not the issue before us today. 



136 

However, in retrospect, I believe that the academic community 
has failed to arm patients, their families and the advocacy groups 
with sufficient information about clinical trials to protect them 
from the overwhelming anxiety that they experienced when they 
were confronted with the news that some data in these trials were 
flawed. I am sure that in collaboration with NIH we can do better. 

We are here because, notwithstanding the soundness of the sci- 
entific conclusions, the quality of administration of NSABP has 
been questioned. In close coordination with the National Cancer In- 
stitute and other cognate agencies, we at the University have been 
working very, very intensively to identify, analyze and respond to 
each of these concerns. 

For example: The University appointed one of its most distin- 
guished senior scientists, Dr. Ronald Herberman, to be interim 
head of NSABP pending the ongoing reviews and reforms. Dr. 
Herberman, who served as an NIH scientist for many years before 
coming to the University of Pittsburgh, agreed to take on this 
heavy and thankless burden, making only the request, which we 
will honor, that as soon as a permanent NSABP chair is in place 
he will return to his duties as head of the Pittsburgh Cancer Insti- 
tute. 

We are committed to finding the best possible candidate to be 
permanent NSABP chair and are now conducting a national search 
for a nationally recognized and distinguished surgical oncologist to 
serve in that role. 

We have proposed and are committed to the implementation of 
a new permanent NSABP leadership structure to ensure that the 
project will be run in a manner that justifies high public con- 
fidence. 

NSABP, under the interim leadership of Dr. Herberman and his 
colleagues, proposed, and the National Cancer Institute accepted — 
after many extensive and substantive interactions — a comprehen- 
sive NSABP plan for corrective actions. Dr. Herberman will de- 
scribe the highlights of that plan to you. I will comment only that 
the plan includes some of the most innovative, reliable and state- 
of-the-art data integrity and audit mechanisms known to science in 
the context of the multi-center clinical trial. 

We have responded openly and cooperatively to the requests of 
this subcommittee and its staff for documents, information and wit- 
ness interviews and have similarly responded to requests from 
what I believe you refer to as the other body. 

As has been reported in the press, the University convened an 
independent panel of nationally recognized experts to determine 
whether an investigation is warranted into possible scientific mis- 
conduct by several NSABP personnel. That panel is hard at work, 
and we are giving them all of the data and support that they need 
and want. 

As you know, Mr. Chairman, and as we have discussed with the 
subcommittee staff, the University is bound by applicable HHS reg- 
ulations to maintain the confidentiality of the misconduct inquiry 
and to preserve the due process rights of the individuals involved, 
who are represented by their own attorneys. 

NSABP, NCI, NIH, HHS and the University of Pittsburgh per- 
sonnel, in response to these important issues, have met and talked 



137 

by telephone innumerable times, exchanged detailed, voluminous 
information and worked continuously to get the facts, analyze the 
issues and meet all concerns. This work is ongoing, but I can report 
that very extensive and positive progress has been made. 

Mr. Chairman, I will tell you frankly that over the past several 
months I have often asked myself what I could have and should 
have done to prevent the concerns that we are addressing today 
about NSABP's administration. That question has taken me down 
many pathways of self-examination, self-interrogation, and some- 
times self-doubt. 

Although I have many flaws, being timid is not one of them. Had 
I been motivated to probe the management of NSABP more deeply, 
I would certainly have done so. Why did I not? 

The answer to this question why did I not is neither easy nor 
clear, and I do not yet have complete confidence in the answer even 
now. The answer, I believe, primarily relates to the culture of def- 
erence that has developed at universities over many, many years, 
if not centuries. The modem research university, although subject 
to strict accountability to government and the public, is primarily 
a highly decentralized system for research and teaching in which 
faculty, and especially well-established senior faculty, have very 
considerable autonomy. 

Are we at the University of Pittsburgh behind the curve in this 
respect? I commissioned an informal survey of 20 leading American 
research universities to find the answer to this question. What I 
learned is that, with the possible exceptions of two universities, 
none of these leading institutions has in place a systematic, peer- 
driven mechanism for reviewing the adequacy of research adminis- 
tration by senior faculty. That finding did not surprise me. 

As I pointed out in an interview some weeks ago to The New 
York Times, the time has come to give very serious consideration 
to changing this culture of deference. This matter requires thor- 
ough examination within the higher education community in gen- 
eral and the biomedical research community in particular. My 
guess is that this culture of deference, as Dr. Broder pointed out 
in his April 13th testimony to the subcommittee, made him hesi- 
tate to call upon the senior academic administration of our Univer- 
sity when NCI did not get a satisfactory response from NSABP. 

I will propose at the University of Pittsburgh, Mr. Chairman, a 
potent mechanism for peer review of the adequacy of administra- 
tion by principal investigators, including senior faculty. We will use 
this new mechanism as a pilot program or model. Dr. O'Connor 
supports our efforts, and I expect that our faculty will support 
them as well when the merit of this initiative is presented. 

Mr. Chairman, I also wish to say a word about research mis- 
conduct. We have taken very aggressive efforts at the University 
of Pittsburgh to deter such misconduct and to encourage the report- 
ing of alleged research misconduct. We have had several cases that 
resulted in a finding of misconduct and on the whole handled them 
to the best of our ability, learning sometimes the hard way from 
our experience. 

I might note that only two of our cases involved fabrication or 
falsification of data included in published research, like the case 
from Montreal that first brought the NSABP matter to our atten- 



138 

tion. It is extremely difficult and perhaps impossible to judge reli- 
ably whether a particular institution has more or less research 
misconduct than another institution. However, common sense sug- 
gests that there will be more misconduct proceedings at institu- 
tions that are alert to misconduct and report it competently, ener- 
getically and efficiently. Generalizations are hazardous about these 
matters and especially about the highly diverse and relatively few 
cases we have experienced at the University of Pittsburgh. 

A legitimate question has also arisen whether it was sound policy 
to accept philanthropic funds from a pharmaceutical company to 
endow a proposed professorial chair when a product of the pharma- 
ceutical company was being researched at the University and by 
the professor in whose honor the chair was to be named. 

On the issue of conflict of interests and the Bernard Fisher-ICI 
Pharma Professorship in Surgery — which to the present day has 
never been fully funded or activated — we simply ask, as has the 
scientific community, that the government's standards in this 
evolving area be made clear and be applied prospectively, not retro- 
actively. Researchers and the institutions at which they work can 
adjust to any reasonable standards, but it is counterproductive to 
judge them by standards not in effect now, let alone years ago. 

Mr. Chairman, like virtually every other research University in 
the United States, the University of Pittsburgh, no less than NIH, 
Congress and the American people, is deeply concerned whenever 
administration of any of its programs falls short of sustained excel- 
lence, whether slightly short or far short. The facts are still not all 
in, but, like the National Cancer Institute, we acknowledge and are 
addressing in detail deficiencies and objectives for improvement. 

The University of Pittsburgh's fundamental aim in relation to 
this matter is neither to defend nor to accuse but to determine dis- 
passionately, fully and as speedily as feasible what happened, what 
can be learned from it and what remedial actions are appropriate 
and proportional to past performance and future commitment. To 
that end, we are hard at work, and we will remain at work until 
we have satisfactory answers. 

Thank you very much, Mr. Chairman. 

Mr. DiNGELL. Thank you. Dr. Detre. 

Dr. Herberman. 

TESTIMO^fY OF RONALD B. HERBERMAN 

Mr. Herberman. Mr. Chairman and members of the subcommit- 
tee, I am Dr. Ronald Herberman, and I am pleased to be here 
today to talk about the National Surgical Adjuvant Breast and 
Bowel Project, the NSABP. 

As interim chairman, I wish to briefly address its past, its 
present and its future. I have provided more detailed remarks for 
you to peruse at your leisure. 

Mr. DiNGELL. Without objection, those will be inserted in the 
record at the appropriate place. 

Mr. Herberman. Thank you, sir. 

First and foremost, my goal as a cancer researcher as well as in- 
terim chairman of the NSABP is to see that this organization re- 
mains viable and effective. The NSABP must continue to develop 



139 

and implement innovative clinical trials, trials designed to advance 
the treatment and prevention of breast and colorectal cancer. 

The NSABP is a most important research program. It needs to 
be preserved and restored to its full vitality. 

Since March 30th, when I was asked by the University of Pitts- 
burgh and by the National Cancer Institute to assume administra- 
tive responsibility for this program, I have felt it my obligation to 
be certain that what the NSABP accomplishes will truly be in the 
national interests. Primarily, it has to operate in the best interests 
of those women and men affected by breast or bowel cancer. In this 
regard, we have taken appropriate steps to move the organization 
forward and to restore confidence in the future of the NSABP. 

Let me tell you about my initial reaction when I assumed respon- 
sibility for the NSABP. I found an organization under siege. There 
was widespread concern. This came from both the NCI and the 
public. It dealt with a variety of problems that the group had not 
adequately handled, but I also knew, as a cancer researcher — in 
fact, director of the Pittsburgh Cancer Institute — that this multi- 
center clinical trials group had a proud history. It had to its credit 
a series of pioneering and even ground-breaking discoveries relat- 
ing to cancers of the breast, colon and rectum. 

My immediate goal was to stabilize the program and to keep it 
on track. I believe that in a short period of only 10 weeks that we 
have been together our new team has accomplished a series of 
major steps towards restoring credibility. Yes, I believe we are get- 
ting the program back on track. 

When I began my remarks I mentioned I would discuss the past, 
present and the future of the NSABP. Let me do that now with 
some specifics. 

Past: You know about the deliberate data falsification problems 
at St. Luc Hospital in Montreal. They have been correctly and well 
documented. 

The present: We are putting a system of data verification in 
place that will minimize the deliberate and also inadvertent errors. 

Data falsification is reprehensible and cannot be tolerated, par- 
ticularly when it may impact on patient welfare. However, in terms 
of effects on the results of clinical trials, the fact is that it doesn^t 
matter if a mistake is deliberate or inadvertent. You must mini- 
mize it from happening. 

Our system is totally unique for monitoring clinical trials at such 
a large number of institutions. 

The future: First and foremost, it will involve the participants in 
verifying information about themselves. When someone volunteers 
to participate in a clinical study, that person will be shown a list 
of dates of key events that will determine eligibility. This might be 
the date of surgery or the date when a mammogram was per- 
formed. 

This procedure is indeed revolutionary. As I just mentioned, 
something like this has never been incorporated into clinical trials 
before. If this had been in place previously, this rather simple step 
would have made it virtually impossible for the problems at St. Luc 
to have occurred. 

So already we have taken the problem of misreporting informa- 
tion, we have addressed it and have come up with a unique, revolu- 



140 

tionary solution. This system, developed in cooperation with Wes- 
tinghouse and Carnegie Group, will have real-time assessment of 
a patient's status and proposed treatment. It will immediately 
bring attention to deviations or inappropriate actions. 

Let me show you this new system briefly in greater detail. In 
other words, we believe that this is the future. It will prevent fal- 
sification of data and prevent inappropriate procedures. 

By the way, this system was unveiled earlier this week at the 
NSABP's annual meeting in Nashville. More than a hundred data 
managers from participating institutions enthusiastically endorsed 
the approach, and many of them were vying for being among the 
sites where the pilots would be carried out. 

Let me read you the comments from one of the data managers 
at this meeting. "We have been requesting software like this for 
years to ensure quality data at our fingertips and for audits, will 
make data collection and data management efficient, will provide 
continuity and quality of care, great job done in 6 weeks." 

Much of the details of this system have been already developed, 
and this has been done in close consultation with cancer survivors. 
We have put together an advisory committee to help us in this re- 
gard, and I will leave for the committee's interest a videotape 
which will show some of the process by which the cancer survivors 
helped us with this important process. 

Mr. DiNGELL. Thank you, Doctor, the committee will receive that, 
and we thank you. 

Mr. Herberman. Thank you, Mr. Chairman. 

Another value of this system is the more complete informing of 
clinical trial participants about their disease, their proposed treat- 
ments and possible consequences. A customized notebook will be 
prepared for each patient electronically which will include informa- 
tion from the NCI's PDQ system and will be tailored for each wom- 
an's individual problems and the proposed treatments so that each 
participant on a clinical trial will be fully informed. 

You will notice that this is in a loose-leaf fashion. This will be 
updated at every visit of the participant to see the doctors and the 
NSABP. 

A third issue deals with the drug tamoxifen. There is a risk in 
taking most drugs. An NSABP study itself has demonstrated a re- 
lationship between the use of tamoxifen for treating breast cancer 
and the development of uterine cancer. There were some delays in 
ascertaining that deaths were a result of uterine cancer. Ways have 
to be found to diagnose uterine cancer early to prevent progression 
and even death. 

Since assuming responsibility we did two major things in this re- 
gard. First, we established an independent data and safety mon- 
itoring question. This meets regularly and reviews all information 
about side effects from any treatments. This group will have the 
power to make appropriate corrective action or even stop the trials 
until problems are solved. 

Second, we have developed a protocol that mandates regular 
uterine examination on all new participants in tamoxifen trials. 
For women already enrolled on a tamoxifen trial, they must at 
least be told about the recommendations for such regular monitor- 
ing. 



141 

By the way, previous NSABP studies have clearly indicated the 
benefits from the use of tamoxifen. The NSABP trials have shown 
that tamoxifen is highly effective in preventing recurrence of the 
original breast cancer. At the same time, the studies demonstrated 
that tamoxifen went a long way in preventing a new cancer in the 
opposite breast, and, in fact, this important finding laid the founda- 
tion for the breast cancer prevention trial that is currently being 
performed by the NSABP. 

In summary, the NSABP has certainly had its share of chal- 
lenges — there is no doubt about that — but things are back on track. 
An effective quality assurance program is already in place. In fact, 
our new system of data verification, I believe, is a major break- 
through. 

The formation of oversight and also data and safety monitoring 
committees add greater credibility than ever before. We have put 
steps into place to streamline headquarters operations and to de- 
velop open communications with patients and with the public. The 
momentum is started so that the NSABP can continue to be the 
high quality organization it was meant to be. I can assure you that 
the NSABP is indeed back on track. 

Thank you very much, Mr. Chairman. 

[Testimony resumes on p. 157.] 

[The prepared statement of Dr. Herberman follows:] 



142 

STATEMENT 
OF 



DR. RONALD B. HERBERMAN 

INTERIM CHAIRMAN 

NATIONAL SURGICAL ADJUVANT BREAST AND BOWEL PROJECT 



Mr. Chairman and members of the Subcommittee, thank you 
for permitting me to appear before you today and provide my views 
as the Interim Chairman of the National Surgical Adjuvant Breast 
and Bowel Project ("NSABP"), a position I have held for slightly 
more than two months. I also welcome the opportunity to answer 
your questions — as I have in meetings with Subcommittee staff 
— and to provide information. 

By way of background and for the Subcommittee's 
information, I also serve as the Hillman Professor of Oncology 
and Professor of Medicine and Pathology at the University of 
Pittsburgh and as Director of the Pittsburgh Cancer Institute. 
The Pittsburgh Cancer Institute is one of twenty-seven federally- 
designated comprehensive cancer centers in the country. It is 
responsible for cancer research and care at the University of 
Pittsburgh Medical Center. Before joining PCI, I served at the 
National Cancer Institute for nineteen years. As a result of my 
work in these positions, I have extensive experience in 
clinically relevant cancer research, clinical trials, and 
research administration. 

I am highly honored to serve as Interim Chairman of 
NSABP, especially at this critically challenging time. Although 
recent events have understandably focused attention on NSABP 's 
past administrative deficiencies, I nevertheless think it 
appropriate to approach the leadership of this organization as an 
act of stewardship over an invaluable national resource that was 



143 



founded and nurtured by the dedication of many others: Dr. 
Bernard Fisher, founder and former Chair of NSABP, who has 
dedicated his entire professional life to eradicating the scourge 
of breast cancer; Dr. Carol Redmond, a renowned statistician who 
has created and managed the quantitative foundation of the 
NSABP 's research; the many members of their staff at the 
University of Pittsburgh and approximately 5000 researchers at 
almost five hundred other facilities participating in NSABP 
trials, who — with some regrettable exceptions — have conducted 
the massive research operation that is NSABP with a high level of 
dedication and professionalism; the federal government and 
taxpayers who provided the financial resources for this endeavor; 
and, most importantly, the 50,000 or so women who have 
participated in the numerous NSABP clinical trials that have so 
markedly contributed to our understanding of appropriate 
treatments for breast cancer. 

Among the major and ground-breaking innovations that 
resulted from the efforts and dedication of these contributors 
are studies that have: 1) provided evidence supporting the use of 
breast conservation surgery as an alternative to radical 
mastectomy, the disfiguring surgery that was previously the 
standard procedure in cases of breast cancer; 2) demonstrated an 
appropriate role for chemotherapy in the treatment of primary 
breast and colon cancer; and 3) demonstrated the value of the 
drug Tamoxifen for the treatment of breast cancer and the 
prevention of second primary cancers. It is on the foundation of 



- 2 - 



144 



these and other significant scientific findings that we now work 
to build a stronger and more accountable NSABP. 

Since assuming the interim chairmanship of NSABP a 
little more than two months ago, I have been engaged more than 
full time in that work. A wide variety of issues and tasks have 
arisen on a number of fronts during that time period. As the 
interim chairman of NSABP, however, I have primarily focused on 
two key priorities. First, one of my main tasks in the last two 
months has been to develop open communications and make concerted 
efforts to restore the confidence of patients and the public in 
NSABP clinical trials and, derivatively, in clinical trials 
generally. Second, and in service of the first priority, I have 
been working, together with Dr. Donald Trump, Interim Executive 
Officer of NSABP as well as Deputy Director of the Pittsburgh 
Cancer Institute, to understand the administrative and other 
problems that exist at NSABP and to respond by developing 
detailed plans to ensure that these problems cannot occur in the 
future. 

Let me first mention generally the types of problems we 
have identified. As made clear by the entire St. Luc episode — 
and several other, less dramatic, episodes that have subsequently 
come to light — NSABP unfortunately suffered from a number of 
deficiencies in quality assurance and also administrative and 
reporting lapses in the face of problems that were detected in 
audits of institutions outside Pittsburgh that enroll patients in 
NSABP studies. On some occasions, for example, questions 

- 3 - 



145 



concerning the eligibility of patients to participate in various 
NSABP clinical trials were identified at the time of an audit but 
not properly resolved. Most of the identified problems may be 
grouped into two categories, one focused on several of the almost 
500 institutions around the United States and Canada 
participating in the clinical trials, and one reflecting 
deficiencies in the NSABP Headquarters: 

1. Deliberate or inadvertent errors in data 
affecting eligibility or other important aspects 
of the studies. Whether such errors are 
deliberate or a result of sloppiness in record 
keeping, they detract from the research and must 
be minimized; and 

2. Slow or ineffective handling of the problems 
once they are detected by Headquarters staff. 

Each such deficiency is of urgent concern. Let me 
stress, however, that nothing we have discovered during this 
period of intense scrutiny has provided any basis to doubt the 
fundamental findings of NSABP 's various ground-breaking studies. 
In fact, various independent clinical trials and also reanalyses 
of the NSABP data — both within and without NSABP — have 
confirmed all of the major conclusions from those studies. 



- 4 - 



146 



In addition, as NCI has known for some time, NSABP had 
in the last year fallen significantly behind schedule in 
performing the audits of institutions. While there may have been 
some plausible reasons for this problem, including the shift to 
implementation during the last year of a new and improved NSABP 
auditing system, we simply cannot tolerate gaps in verification 
of information and practices at participating institutions. In 
light of the particular concerns that have been expressed 
concerning the gathering and handling of information relating the 
use of Tamoxifen to increased risk of developing endometrial 
cancer, we have also been re-examining intensively the entire 
NSABP system for identifying and reporting adverse drug reactions 
that may be experienced by patients enrolled in NSABP protocols. 

These are some of the types of problems we have 
identified in our efforts to make NSABP a stronger and more 
accountable research organization. In response to these problems 
and others, however — and in the midst of crushing demands for 
information from various government entities and the press — we 
have in the space of two short months implemented a number of 
significant structural and administrative changes that we believe 
will both address these problems directly and restore the 
confidence of participants in NSABP clinical trials, the breast . 
and bowel cancer community, the scientific community generally, 
and the public at large. These actions and proposals for further 
action are explained in detail in the "NSABP Plan for Corrective 
Actions" that was submitted to NCI on May 24. A copy of this 



- 5 - 



147 



Plan has been provided to the Committee for its information. I 
summarize here some of the key components of this plan. 

Working closely with Dr. Samuel Broder, Director of 
NCI, Dr. Bruce Chabner, Director of NCI's Division of Cancer 
Treatment, and other NCI officials, we have established the NSABP 
Oversight Cconmlttee as an external advisory group. Members of 
the Committee include experts in large-scale, clinical cancer 
trials. Of particular significance, moreover, the Committee 
includes two lay members who serve as advocates for breast cancer 
patients and survivors: Amy Langer, Executive Director of the 
National Alliance of Breast Cancer Organizations, a member of the 
Board of Directors of the National Breast Cancer Coalition, and a 
breast cancer survivor; and Dorothy Raizman, an attorney who is a 
recognized authority on medical records and the communication of 
information within and throughout the medical system. The NSABP 
Oversight Committee is advising me on reorganization of the NSABP 
leadership structure, and will continue in place at least until 
the NSABP is stabilized and functioning more effectively. 

With the assistance of the NSABP Oversight Committee, 
we have proposed a plan for a new NSABP leadership structure. We 
are working on revisions of the NSABP constitution and bylaws 
that will shift a good deal of responsibility for management of 
NSABP from the group Chair to a reconstituted Executive 
Committee. The Executive Committee membership will be more 
broadly representative of all group constituencies than the 
current Executive Committee. Following reconstitution of the 

- 6 - 



148 



Executive Committee, that body will nominate candidates for a new 
permanent group Chairman, and this important step will be 
followed by an election by the membership of the group, and 
approval of the selection by the National Cancer Institute. 

A major goal of our revised leadership structure is to 
ensure greater participation by the health care providers and 
researchers In local NSABP centers In the design and execution of 
research protocols . I hasten to note that, even under the former 
management structure, protocol design committees and the NSABP 
group meetings have generally provided excellent opportunities 
for group-wide participation in protocol design. Our proposed 
management structure will further enhance membership involvement 
by establishing standing disease-specific (breast or colorectal) 
committees, each with a chair empowered to review continuously 
the status of ongoing trials, solicit group-wide and ad hoc 
external input, and guide development of new protocol concepts. 
These committees will be responsible to the Executive Committee, 
which, as noted, will itself have broader membership and more 
inclusive group representation than under its current structure. 

We have also established a Data and Safety Monitoring 
Committee, the membership of which is completely independent of 
NSABP. The primary responsibility of this Committee is to review 
interim analyses of outcome data and to recommended changes in — 
or termination of — studies based on these analyses. The 
Committee will also review toxicity data, major modifications of 
studies, NSABP handling of scientific misconduct, NSABP quality 

- 7 - 



149 



assurance, and consent form issues related to toxicities. 
Included as a member of this Committee is Kay Dickersin, Ph.D., 
Department of Ophthalmology, University of Maryland, who is a 
breast cancer advocate and survivor, with considerable expertise 
in clinical trials and epidemiology. 

In consultation with NCI and experienced auditors from 
other cooperative groups, we have also developed an improved on- 
site monitoring and quality assurance program. Under these aaw 
audit procedures, an institution will be notified four weeks in 
advance of a proposed audit, and will be informed two weeks in 
advance as to which cases are to be audited. These cases are 
randomly selected, and additional cases are identified on site at 
the time of the audit. Document review will occur on site, 
rather than at NSABP as under the former system. Each audit team 
will include an independent member and NSABP group members will 
be involved in the actual performance of the audits. To assist 
in the implementation of these new procedures, we have 
subcontracted with a consulting firm experienced in the auditing 
of oncology trials. 

NSABP will notify the Clinical Trials Monitoring Branch 
("CTMB") of NCI of any major deviation within 24 hours of 
discovery, and will send by facsimile the preliminary audit 
finding within 24 hours of completion of the audit. Within seven 
days of the auditors' return and after review by the Executive 
Officer, NSABP will transmit an audit summary to NCI and the 
audited institution. The institution will then be given 20 days 

- 8 - 



150 



to respond. Within 30 days of the preliminary report and after 
approval by the Quality Assurance Committee, a final report 
containing the Executive Officer's recommendations will be sent 
to NCI and the institution. All institutions will be at risk of 
annual audits, and will be audited at least once every three 
years . 

NSABP has proposed a schedule that will eliminate the 
back-log of audits by April 1, 1995, with highest priority given 
to institutions with large numbers of accruals, previously 
recognized problems, or a gap of more than three years since the 
previous audit. A review of past audits will be used to identify 
outstanding issues and priorities for future audits. NSABP will 
provide NCI semi-annually an updated list of institutions to be 
audited within the next six months. New procedures are being 
developed to provide for timely auditing of low-accruing 
institutions and small affiliates. 

In response to problems identified in past audits, 
NSABP has assisted NCI in organizing re-audits, followed up on 
identified deficiencies, and taken significant corrective 
measures. In addition, we have initiated a broad scale review of 
all previously conducted audits — and NSABP 's handling of 
deficiencies identified by those audits — to further ensure that 
all questions involving quality assurance are appropriately 
resolved. 



- 9 - 



151 



We have also been examining ways to increase the 
reliability of information at the time it is initially entered 
into the NSABP process. Calling upon existing relationships with 
the Westinghouse Electric Corporation's Science and Technology 
Center and Carnegie Group, Inc. Dr. Trump and I have worked to 
generate new processes for handling data monitoring and quality 
control issues. I should note that we began to develop these 
processes over a year ago, well before the current controversy 
over NSABP arose. One important new procedure will involve 
patients — who likely are in the best position to verify certain 
information — in quality control procedures incorporating 
important patient history dates into the consent form for patient 
verification. The University of Pittsburgh has also been working 
with Westinghouse and the Carnegie Group to develop a prototype 
system for data confirmation and quality control, which was 
demonstrated earlier this week to investigators and data managers 
at the NSABP group meeting in Nashville. Improvement of this 
information process continues, and a pilot test and verification 
of its effectiveness are planned. 

One of our primary concerns is to Increase the 
invol-vemeat of breast and colorectal cancer survivors and their 
advocates in the NSABP. A comprehensive plan to achieve this 
goal is being developed and will be presented to NSABP members at 
our group meeting later this month. Under this innovative plan, 
representatives of these groups will: 1) attend and participate 
in each NSABP membership meeting; 2) serve on protocol 



- 10 



152 



development committees; 3) serve on focus groups to provide input 
on protocol design, informed consent documents, and audit 
procedures; and 4) participate in the development of plans for 
assessing such issues as effects of clinical trials on quality of 
life, compliance of patients on clinical protocols, and 
psychological aspects of clinical trials. We will also develop a 
procedure to inform participants in a trial and their families 
about the results of a trial as they are published. We hope to 
establish a close and ongoing partnership between patients and 
their physicians, and to ensure that patients are fully educated 
about the clinical trials in which they participate. 

Let me also mention some specific actions we have taken 
in responso to St. Luc falsifications and the questions about 
whether those were adequately disclosed. NSABP officials 
previously had determined (in July 1991) that the falsified data 
did not affect study outcomes, and had given a presentation to 
NCI officials in March 1992 on their statistical reanalysis of 
data from four NSABP trials, excluding falsified data from St. 
Luc Hospital. Although NCI and ORI had asked NSABP to prepare a 
reanalysis for publication immediately following the conclusion 
of OKI's investigation into St. Luc, the reanalysis was not 
assigned the priority that it merited. NSABP submitted the 
formal manuscript to NCI on February 8, 1994. 

On March 17, 1994 Dr. Fisher notified the New England 
Journal of Medicine ("NEJM") that papers published in that 
journal concerning three NSABP trials contained falsified data 

- 11 - 



153 



from St. Luc. NSABP had mistakenly decided not to notify NEJM of 
the ORI findings earlier because those findings had been 
published in the Federal Register, because Drs. Fisher and 
Redmond were preparing a formal paper for submission to NEJM, and 
because the falsified data did not affect the study results. The 
University acknowledged that this was a mistake in a March 28, 
1994 letter to the editor of NEJM. On March 25, after NCI's 
review, the paper reanalyzing data from the three trials was 
submitted to NEJM for publication. 

On March 30, Dr. Fisher notified in writing all other 
journals that had published articles including fraudulent St. Luc 
data since 1991. NSABP withdrew manuscripts including tainted 
data that had been submitted or were in press. On April 7, 1994, 
we sent all principal investigators a summary of ORI ' s findings 
regarding St. Luc and a model letter to patients reassuring them 
of the validity of the study results and acknowledging the 
misperception creatied by NSABP 's delay in publicizing the 
problem. On May 6, 1994 I wrote to the editors of all affected 
journals and affirmed NSABP 's commitment to reanalyzing the trial 
data and confirming the studies' validity. 

NSABP currently is preparing a manuscript describing 
the reanalysis of all affected trials. This reanalysis will be 
provided for NCI's review by June 30, 1994 and submitted for 
publication upon NCI's approval. A detailed description of the 
reanalysis appeared in a Technical Report distributed at our 
NSABP meeting (which was held over the last several days) and is 

- 12 - 



154 



being made available to journals that have published reports 
including St. Luc data. Each if these journals will also receive 
notices of the formal, published reanalysis. At our recent NSABP 
meeting, we had an extensive discussion of the St. Luc incident, 
which I believe served to educate NSABP physicians, data 
monitors, and support staff about the St. Luc problem, the 
ethical and legal consequences of scientific fraud, and NSABP' s 
new procedures for preventing and detecting data falsification. 
A letter describing the St. Luc problem has been developed in 
consultation with breast cancer advocates and sent to a wide 
range of individuals and organizations. 

With respect to problems at other institutions, I think 
it is important to stress that none of the other cases we are 
examining have been determined to involve the type of deliberate 
and systemic falsifications that were identified at St. Luc, 
although investigation is ongoing. That does not, of course, 
diminish in the least the need to address data integrity issues, 
for we should strive to eliminate all error — whether deliberate 
or inadvertent — from our data set. 

In connection with the discovery of evidence of 
possible scientific misconduct at St. Mary's Hospital in 
Montreal, NSABP has worked to correct the problems leading to a 
delay in reporting that evidence to NCI, and has provided NCI 
with information and assistance in re-auditing St. Mary's. ORI's 
investigation into evidence of scientific misconduct at St. 
Mary's is expected to conclude in the near future. 

- 13 - 



155 



Administrative problems that have been identified 
during audits of Louisiana State University ("LSU") and Tulane 
University led to suspension of these institutions that month, 
prior to the general suspensions of NSABP institutions. These 
problems have been identified to the principal investigators at 
LSU and Tulane, whose responses and proposals for solving the 
problems will be addressed. Dr. Trump as Interim Executive 
Officer is reviewing these responses and together with NSABP 
staff will determine whether accrual at these sites should be 
resumed. We have also been working with NCI to address problems 
relating to Memorial Cancer Research Foundation in Los Angeles 
and its principal investigator. Dr. David Plotkin, and South 
Nassau Community Hospital. 

We are well aware that NCI has perceived an 
unresponsiveness or inadequate responsiveness from NSABP in the 
past, particularly in the administrative realm. That perception, 
whatever the basis for it, is destructive of the close working 
relationship with the funding agency that is necessary to the 
conduct of federally sponsored research. We have taken very 
substantial and concrete steps to ensure responsiveness to NCI, 
and to prevent such a perception henceforth. 

I should also mention, as has been reported in the 
press, that the University has initiated a scientific misconduct 
inquiry relating to events and actions at NSABP. A highly 
distinguished three-njember panel has been convened. Each of the 
panelists is a nationally recognized authority, based at another 

- 14 - 



156 



institution, holds no position at the University of Pittsburgh, 
and is independent of both the University and NSABP. The purpose 
of the preliminary inquiry is to determine whether a formal 
investigation of this matter as a scientific misconduct case is 
warranted. Fairness to the respondents in that inquiry compels 
me to refrain from further comment. 

As I said at the outset, I view my service to NSABP as 
an act of stewardship over a valuable national resource. It is 
my firm belief that the actions I have described, and the path we 
are now on, will not only preserve that resource from 
deterioration, but will also nurture and rejuvenate NSABP so that 
it may continue and expand its service to the women of this 
nation for years to come. I am pleased that we have made such 
significant progress that we were permitted last week to announce 
that NSABP will begin to accrue patients once again into its 
vitally important clinical trials. I look forward to working 
with the Members of this Subcommittee, NIH, NCI, and breast and 
bowel cancer patients and their advocates, to ensure that this 
step is only the initial evidence of the successful 
reorganization of the NSABP. 

I would be pleased to answer any questions you may 
have. 



- 15 - 



157 

Mr. DiNGELL. Gentlemen, the Chair wants to thank you for your 
very fine statements. 

Dr. O'Connor, as you indicated, the committee has sharp teeth, 
and we don't always like to use them on people, but occasionally 
these things happen. I want to commend you and you, Dr. Detre, 
and you, Dr. Herberman, for, first, fine statements; second of all, 
to tell you that the staff and I have appreciated the way that you 
have worked with us and we have worked with you to address the 
concerns of the committee and to tell you that, from your state- 
ments and from what we have seen as we have worked moving to- 
wards these hearings, my belief that the process of scientific review 
and supervision and auditing will move much forward because of 
the good work which you have done and are doing, and we com- 
mend you for that. Thank you. 

Mr. O'Connor. Thank you. 

Mr. DiNGELL. The Chair does have some questions, and the 
Chair advises that the Chair will be recognizing members for 10 
minutes rather than 5 in order to address the questions that we 
have before us. 

I would like to observe to you. Dr. Detre, that I come from a part 
of the country where accents are common. And we are not only 
comfortable with them, but we think that our best and most pro- 
ductive people very frequently have accents. And we are com- 
fortable with them, and we rejoice that we have people who have 
accents who can make meaningful and good contributions. So I am 
comfortable with it, and I hope you are, too. 

Mr. Detre. Thank you, sir. 

Mr. DiNGELL. First of all. Dr. Detre, since this situation unfolded 
the University of Pittsburgh has initiated a number of separate re- 
views or investigations which have been referred to in the com- 
ments of yourself, Dr. O'Connor and Dr. Herberman. Without dis- 
cussing the questions of scientific misconduct in that investigation 
that is currently under way, could you describe in general what you 
found regarding the management and administration of NSABP? 

Mr. Detre. Yes, Mr. Chairman, that is a very important ques- 
tion. Literally every component of the NSABP functions reasonably 
well, but the coordination among the various functions showed con- 
siderable deficiencies. This, in turn, led to delayed reporting in 
some instances or lack of reporting in others. 

Mr. DiNGELL. Now, Dr. Detre, what would you say went wrong 
with management and administration of NSABP? 

Mr. Detre. Mr. Chairman, I don't think I can be more precise 
about it except to say that the administration of NSABP needs 
some reform. Reform which would ensure, as Dr. Herberman point- 
ed out, that the database for clinical research begins with the pa- 
tient, that there is a clear trail following the data through the var- 
ious phases of the treatment trial and that the reporting relation- 
ships are clearly defined and that an independent mechanism is in 
place to ensure that not just the audits are done but audit reports 
arrive through the proper agencies at the proper time. 

Mr. DiNGELL. Dr. Herberman, we understand that your own in- 
vestigations have revealed that NSABP was violating its own con- 
stitution with respect to the recruitment and the selection of par- 
ticipating sites. Could you please describe what you found in this 



158 

regard? Specifically, how was this process supposed to work, and 
then how did it, in fact, work? 

Mr. Herberman. The constitution that currently governs the 
NSABP is a rather old document. It is dated 1969 and, as best as 
we can determine, has had very little updating since then. We are 
currently working on that very matter and will soon have a new 
and revised constitution. 

The description in the constitution, as the original document 
which is now undergoing extensive revision, was that member in- 
stitutions would need to be voted on by the membership of the 
NSABP. For the last number of years, at least, this has not been 
the practice, but rather there was an administrative mechanism 
within the headquarters operation for one of the staff to collect all 
of the necessary documents and to essentially decide whether all 
the documents were in place and, if so, to approve an investigator 
or even an institution for membership in the NSABP. 

Mr. DiNGELL. Now, Dr. Herberman, what is your own view of the 
effect that this violation of the NSABP's own constitution had on 
the selection of sites and their ability to produce verifiable data? 

Mr. Herberman. Well, the NSABP has grown to be a very large 
and complex organization. It has almost 500 sites which vary in 
their nature and their quality. I believe that processes need to be 
in place to ensure the training and the quality of each of the insti- 
tutions that participate and procedures are now being put into 
place to be sure that the smaller institutions and the ones that are 
likely to accrue small numbers of patients will affiliate themselves 
with larger institutions in their region so that there can be a direct 
help to be sure that the proper procedures are always carried out. 

Mr. DiNGELL. Now, Dr. Herberman, can you tell us what you 
have learned about the methods by which NSABP reimbursed its 
sites for patient recruitment? Specifically, were the majority of the 
sites paid for accrual patient by patient? 

Mr. Herberman. There have been a variety of mechanisms, Mr. 
Chairman, for reimbursing institutions that participate in the clini- 
cal trials. There currently are 11 institutions that receive grants di- 
rectly from the National Cancer Institute for their participation in 
NSABP clinical trials. 

The much more common mechanism at the moment is the dis- 
bursement of money through purchase service agreements that is 
administered by the headquarters at the University of Pittsburgh. 
For the purchase service agreements, the mechanism essentially is 
a per capita payment for each participant entering onto a clinical 
trial. 

Mr. DiNGELL. Now, Dr. Herberman, is it not also true that for 
those sites payments were routinely made before eligibility for the 
patient had been confirmed and long before the patient entered fol- 
low-up? 

Mr. Herberman. Well, sir, my understanding has been that the 
practice is that a trigger for payment for a participant would occur 
at the time of randomization. A patient would not be randomized 
until checks for eligibility had been completed. But if there subse- 
quently, after randomization, was a determination of ineligibility, 
there was not a mechanism in place to recover such funds. There 
also was not a mechanism for adjusting the funds relating to the 



159 

extent or the quality of the follow-up. The payment was made at 
the beginning, with the presumption being that this was funding 
for the entire process, the entry and the follow-up of each partici- 
pant. 

Mr. DiNGELL. Two questions, then. Did this contribute to eligi- 
bility discrepancies? And did it contribute to follow-up delin- 
quencies? 

Mr. Herberman. It is hard to determine an answer to that, Mr. 
Chairman. I believe that the very vast majority of the participating 
institutions have been extremely conscientious in carrying out all 
aspects of the clinical trials that they participate in. 

It is also important to note that the amount of funding that was 
coming from the Federal Government did not provide full funding 
for all of the aspects of the clinical trials that institutions partici- 
pated in. I am convinced that the primary motive for institutions 
and the physicians associated with them to participate was their 
enthusiasm about the important work that was being carried out 
and not directly linked to the payment mechanism. 

Mr. DiNGELL. Now, Dr. Detre, did NSABP ever attempt to recap- 
ture funds for patients ultimately proven to be ineligible or un- 
available for follow-up? 

Mr. Detre. Not to the best of my knowledge, Mr. Chairman, but 
I may be in error. 

Mr. DiNGELL. Could you give us — could you tell us why? 

Mr. Herberman. Mr. Chairman, if you would permit me, I think 
I might answer that question. This came up for some discussion 
when the National Cancer Institute officials visited the University 
of Pittsburgh last month. 

What was pointed out was that, because only partial reimburse- 
ment for the total costs of the trials were being performed, there 
was a general feeling that the amount of money that was being dis- 
bursed was still less than the total cost that an institution would 
have, particularly because the proportion of ineligible patients at 
any given institution has tended to be relatively low. 

Mr. DiNGELL. Thank you, doctor. 

The Chair notes that I have used about all the time that I should 
at this particular time so the Chair is going to recognize my good 
friend from Colorado, Mr. Schaefer. 

Mr. Schaefer. Thank you, Mr. Chairman. 

I would just like to, at the outset, reflect what the chairman stat- 
ed to you gentlemen about the cooperation with our subcommittee 
stafY in trying to find all the answers that we and the American 
public really needs to know on this matter. 

Earlier this week. Dr. O'Connor, the NCI announced it was open- 
ing competition for the management of NSABP. Is the University 
of Pittsburgh going to compete in this? Up to this point it had just 
been a given, as I understand it. 

Mr. O'Connor. We certainly do intend to compete. Congressman 
Schaefer. 

Mr. Schaefer. Now, if you do compete in this, what specific ap- 
proach would you take that differs from the past management of 
NSABP? 

Mr. O'Connor. I think Dr. Herberman will be in a better posi- 
tion to fill in some of those details, Congressman, but I will say 



160 

that we have aggressively pursued an alteration of the organization 
so that it will be both more effective in monitoring and in commu- 
nicating when that needs to happen. But I think Dr. Herberman 
is probably in a better position. 

Mr. SCHAEFER. Doctor, I know both of you have mentioned 
changes in your testimony, but do you want to amplify a little 
about it? 

Mr. Herberman. Yes. Thank you for the opportunity to respond, 
Congressman Schaefer. 

As I briefly described in my testimony, we have very rapidly put 
into place an extensive series of corrective steps and I believe im- 
portant improvements in the way that the NSABP will operate. Be- 
cause of this intense focus on these problems and giving virtually 
full-time dedication not only on my part and Dr. Trump's part and 
several other staff members, I believe that we are now uniquely 
suited to be able to stabilize the group and to keep its important 
programs on track. 

We certainly intend to continue this progress and particularly 
with the development of these innovative data management and 
quality assurance programs. We believe that this is the best pros- 
pect for the NSABP to function at the highest possible level. 

Mr. Schaefer. It seems to me that in the past, with the NCI 
automatically having the University of Pittsburgh do it, there 
might have been some complacency out there and maybe this con- 
tributed to what occurred. I applaud you for taking some new 
steps. 

What, may I ask, would Dr. Fisher's role play if the University 
of Pittsburgh were to get this particular grant? 

Mr. Herberman. Congressman Schaefer, the one thing that is 
clear, I believe, is that there is no plan for Dr. Fisher to resume 
administrative responsibilities for the NSABP or to have direct-line 
authority for any aspects of the program. 

Mr. Schaefer. Do the rest of you gentlemen concur with this? 

Mr. O'Connor. That is correct, Mr. Schaefer. Dr. Fisher has done 
some heroic research in the past and will probably continue to be 
able to do that, but he will not have administrative responsibility 
in NSABP. 

Mr. Schaefer. You stated, and I find it reassuring, that 
NSABP's research findings continue to be sound, and charges of de- 
ficient administration are not adequately answered by merely as- 
serting that the research remains valid. I gather that then you 
would dispute those who characterize the problems at NSABP as 
trivial or nonevents? 

Mr. O'Connor. Congressman, I don't think that falsification of 
anything is ever acceptable and particularly when it deals with the 
public health. 

Mr. Schaefer. So it would not be trivial? 

Mr. O'Connor. It is not trivial, sir. 

Mr. Schaefer. Thank you. 

Mr. Herberman, I am encouraged in your testimony concerning 
the steps you are taking to improve the quality control system of 
NSABP. It is, of course, prospective in nature. What plans do you 
have to address the backlog of audit reports and ensure that the 
information currently in NSABP's possession is accurate? 



161 

Mr. Herberman. Congressman, this is a matter of great concern 
to us. Verification and assurance of the quality of the data, I be- 
lieve, is of primary concern. We are working assiduously to develop 
an audit schedule in which all of the institutions that had not been 
audited during the requisite 3-year period will be audited within 
the very near future. 

In addition, all of the institutions in which we have detected 
problems in previous audits will be given the highest priority for 
a reaudit and more attention. In order to be sure that this process 
moves as expeditiously as possible and to catch up, we have been 
making arrangements with a contractor organization with experi- 
ence in clinical trials related to cancer to help us with this matter, 
at least until we can catch up in the backlog. 

Mr. SCHAEFER. Two questions. How many of these programs 
have you identified as being problems? And have you any idea 
what this is going to cost? 

Mr. Herberman. One of the things that we commissioned shortly 
after I took on responsibility for the NSABP was to go back over 
the last 4 years of all the audit reports. This was over 120 audits. 
Among those we have determined there are about 11 that had sig- 
nificant problems that might affect eligibility or have important im- 
pact on the clinical trials. These will be the ones we will focus on 
particularly. 

Mr. Schaefer. That is a pretty high percent. We are talking 9, 
10 percent. 

Mr. Herberman. It was about 10 percent. These are serious con- 
cerns, and we will deal with these expeditiously. 

Mr. Schaefer. Again, you don't have any idea of what the cost 
is going to be involved in this? You may not have at this point, and 
I understand that. 

Mr. Herberman. I really don't have figures on the costs at my 
fingertips. 

Mr. Schaefer. And I understand that. 

You are the interim chairman of NSABP. What is the time frame 
on the permanent chairman? Will that be this fall sometime? 

Mr. O'Connor. Congressman Schaefer, if I may, we have just es- 
tablished a search committee of approximately five people, and I 
am going to chair that search, and we hope to have the individual 
identified by late summer, early fall. 

Mr. Schaefer. Good. Dr. Detre, do you believe the peer review 
system is adequate when it comes to administration oversight? 

Mr. Detre. Would you repeat the question. Congressman Schae- 
fer? 

Mr. Schaefer. I would be happy to. Do you believe that a peer 
review system is adequate when it comes to administration over- 
sight? 

Mr. Detre. I don't believe so. Congressman Schaefer. I believe 
that there must be some additional institutionally based mecha- 
nism to guarantee that the administration is adequate for the task 
at hand, particularly in such large-scale clinical trials as the 
NSABP which over the years literally included tens of thousands 
of patients at 400 or 500 different sites. 



162 

Mr. SCHAEFER. In advocating the guidelines when it comes to ac- 
cepting philanthropic funds from pharmaceutical companies, are 
you advocating setting up Federal guidelines for this? 

Mr. Detre. Mr. Chairman, our conflict of interest guidelines are 
in place at the University of Pittsburgh. At this point — and that I 
believe is the national standard — we require our faculty to report 
any funds that come from pharmaceutical companies. 

Mr. SCHAEFER. Mr. Chairman, I would yield back. 

Mr. Brown [presiding]. Thank you, Mr. Schaefer. 

Dr. Herberman, we have reviewed your May 12th of this year 
draft report of your review of NSABP recent audits for several 
dozen sites with major discrepancies. There was no indication that 
those discrepancies had been resolved or that the potential irreg- 
ularities further examined. How did this occur? Why was this oc- 
curring at NSABP? 

Mr. Herberman. I think that is an important question. Con- 
gressman Brown. I really don't have a complete answer to what the 
reason for these lapses might have been. We are very concerned 
about it, and I can assure you that now that I am aware that 
these — some of these reporting loops are open, we are taking steps 
to immediately close them. 

Mr. Brown. What do you intend to do? What are you going to 
do to follow up on these reports? 

Mr. Herberman. Well, because we were not entirely certain 
which of some of these reports had been communicated to the NCI, 
we have sent all of them to the National Cancer Institute so they 
have been fully informed. 

In addition to that, we are in the process of notifying each of the 
principal investigators related to these institutions. It is, frankly, 
difficult for us to determine from the files which principal inves- 
tigators were informed and which were not informed about the 
problems. And to ensure that everyone who needs to know is in- 
formed, we will inform all of them. 

Mr. Brown. Why weren't these principal investigators informed? 

Mr. Herberman. Again, Congressman Brown, it is not clear to 
me what was the basis for the lapses in the past. I, frankly, have 
focused my attention not so much on determining the why but the 
nature of the problem itself and what needed to be done for correc- 
tive action, and this is what I believe we have done rather thor- 
oughly in the last 10 weeks. 

Mr. Brown. NSABP's own audit report cites the possibility that 
patients have been given incorrect or inappropriate treatments for 
medical conditions or their care has otherwise been altered or com- 
promised. What responsibility does NSABP have to look into the 
well-being of these women? 

Mr. Herberman. Well, Congressman Brown, this is an issue of 
greatest concern to us. Any actions or inactions that would affect 
the welfare of patients is, of course, of first priority and concern. 
It is difficult in many cases well after the fact when an audit is 
performed, which is often 2 or more years after a patient has been 
treated, to take a really effective corrective action for that particu- 
lar woman. This is one of the reasons why we are putting this new 
system in place, so that we can get immediate, real-time assess- 



163 

ment of problems and correct them while the woman is still under 
treatment so the appropriate treatments can be provided. 

Mr. Brown. Dr. Detre, do you have anything to add to that? 

Mr. Detre. I am not, of course, an oncologist, Mr. Chairman, 
Congressman Brown. I am just an ordinary physician who is a 
neuropsychiatrist by training. 

I would say that in those instances that the questions about eli- 
gibility for trial for inappropriateness of treatment, NSABP under 
a new organization will provide for the reexamination of these 
women to provide them with the best possible advice. 

I might add that while there may have been problems on eligi- 
bility, which is inexcusable, or treatment, which is inexcusable, it 
does not necessarily mean that there are serious consequences of 
these deficiencies. These can be determined only by thorough exam- 
ination of the patients and proper consultation and advice to them. 

Mr. Brown. When will you have done that? 

Mr. Detre. As soon as all of these problems are identified. 

And that process is ongoing — Dr. Herberman probably will tell 
you when all these audits are finished. Individuals who if for any 
reason whatsoever may not have received proper care or were en- 
rolled in trials for which they were not eligible have to be examined 
by experts and not on site but independent experts. 

Mr. Brown. Dr. Herberman, tell us, if you would, if you would 
identify the sites which had the most significant audit discrep- 
ancies. Would you identify those for us? 

Mr. Herberman. Certainly. I think the sites that had the most 
significant problems in fact are the ones that have come to the pub- 
lic attention. In addition to St. Luc Hospital in Montreal there is 
a second hospital in Montreal, St. Mary's, that my understanding 
from Dr. Broder there is a determination of some deliberate alter- 
ation which, again, is of grave concern to us, particularly as it has 
occurred in the prevention trial. 

Beyond that, the institution where there have been, I believe, the 
most serious concerns or allegations brought forward has been at 
Memorial Cancer Research Foundation in Los Angeles. In addition, 
there have been a few other sites in which there have been mul- 
tiple instances of what I would characterize as sloppiness in the 
handling of the records and the clinical trials data, and for those 
this would include particularly LSU and Tulane in New Orleans 
and South Nassau Communities Hospital in Long Island. 

Mr. Brown. The Los Angeles site — apparently, there was an 
unmailed audit report. Can you tell us about that, the United Me- 
morial Cancer Foundation? 

Mr. Herberman. We first learned about the issue at Memorial 
in Los Angeles after it was reported in the newspaper and when 
Dr. Plotkin, the principal investigator there, requested the Na- 
tional Cancer Institute to do an audit. I spoke immediately both be- 
fore the audit and while the audit was going on with NCI officials 
and determined that there were serious enough problems to place 
the principal investigator. Dr. Plotkin, on — suspend him from his 
participation in the clinical trials. 

Why this prior audit report which described a number of discrep- 
ancies in 1990 was not mailed is very difficult to determine. As 
best as I can surmise, the intention was to perform a reaudit and 



164 

to provide that letter about the prior problems to Dr. Plotkin at 
that time, but that reaudit was not performed. 

Mr. Brown. You had earlier said there were 11 sites. You men- 
tioned, I believe, 6. You mentioned two in Montreal, two in Louisi- 
ana, one on Long Island and the Los Angeles site. What are the 
other five? 

Mr. Herberman. I don't have the list of the others in front of me, 
but there are several others. We could provide this to the commit- 
tee almost immediately. 

Mr. Brown. Would you please submit that to us? Is one of those 
Pittsburgh? 
Mr. Herberman. No, sir. 

Mr. Brown. What is the nature of the most serious discrepancies 
that you found in the audit reports? Tell me a little bit about the 
nature of them. 

Mr. Herberman. Well, the type of discrepancies or problems 
vary rather widely. The most serious one I would characterize re- 
lates to information that would affect eligibility of patients entering 
onto a trial. 

A second area of considerable concern has been the documenta- 
tion of the informed consent and exactly when the informed con- 
sent was administered and signed by the patient. 

In addition to that, there have been some questions or possible 
errors related to laboratory values of patients and some other is- 
sues about follow-up, but most of it, I would say, was in the area 
of eligibility and consenting. 

Mr. Brown. Dr. O'Connor, for a moment, you had stated in your 
opening statement, "I accept responsibility for past administrative 
shortcomings and deficiencies of the NSABP project, which is 
headquartered at the University." Tell the subcommittee exactly 
what you mean by that, "accepting responsibility." 

Mr. O'Connor. I will be happy to do so. Congressman Brown. 
I accept the responsibility of putting together an administrative 
team with Dr. Detre and Dr. Herberman which will review the ad- 
ministration of the NSABP grant and bring it up to an administra- 
tive quality that is demanded by the kind of work that it under- 
takes. 

Mr. Brown. The situation is going to be costly in at least three 
different ways. One is the cost to the government for investigating 
and auditing this whole situation; second is the cost to the Univer- 
sity for conducting its various reviews and inquiries, and your com- 
ing here and all that; and third is the cost of recruiting patients 
whose data was found, for a variety of reasons, to be unusable. 

Does Pittsburgh intend to reimburse the Federal Government for 
the cost incurred in this whole investigation of this matter? 

Mr. O'Connor. It certainly is a point that is under discussion, 
Congressman Brown, and I think it will continue to be under dis- 
cussion. 

Mr. Brown. Under discussion among whom? 
Mr. O'Connor. Well, it certainly has been a discussion with NCI. 
Mr. Brown. You will make the final decision, is my understand- 
ing, at the University of Pittsburgh on whether or not the Univer- 
sity agrees without any request or action from us, from the Federal 



165 

Government, on reimbursement. What are your thoughts on that 
today? 

Mr. O'Connor. That I would Hke to keep the discussion going, 
sir. 

Mr. Brown. You would like to keep the discussion going. 

What is your personal — if you had to make a recommendation to 
your — since you are making the decision, if you had to make the 
recommendation to your board of trustees today, would you rec- 
ommend a specific amount that you reimburse the government, 
taxpayers, anything, that you reimburse the full amount? 

Mr. O'Connor. Congressman Brown, I don't have a specific fig- 
ure in my head at this point. That is why I would like to keep the 
conversation going, so that I can 

Mr. Brown. Do you think you owe us something back in dollar 
amounts — not specific dollar amounts, but do you think you owe a 
dollar amount to us, back? 

Mr. O'Connor. Congressman, I need to examine that very care- 
fully, very carefully. 

Mr. Brown. You are in charge, right? 

Mr. O'Connor. That is correct. 

Mr. Brown. Can you assure the subcommittee as a matter of pol- 
icy that Pittsburgh will characterize and categorize all the costs as- 
sociated with this situation as unallowable for Federal reimburse- 
ment? 

Mr. O'Connor. Excuse me, I missed the beginning part of the 
question. 

Mr. Brown. Can you assure the subcommittee that as a matter 
of policy that Pittsburgh will characterize and categorize all its 
costs associated with this situation as unallowable for Federal re- 
imbursement? In other words, the costs that you are bearing, you 
will — the law firms and whomever you may have retained or hired 
or expended moneys upon, that you will not ask any of that for 
Federal reimbursement? 

Mr. O'Connor. That is correct, sir. 

Mr. Brown. OK. Does the University of Pittsburgh plan to reim- 
burse the Federal Government for costs associated with generating 
significant amounts of unusable data? Putting aside the series of 
questions a moment ago, do you plan to reimburse the government 
for generating that data that was unusable? Should taxpayers pay 
for that? 

Mr. O'Connor. We have not come to a conclusion on that yet, 
Congressman Brown. 

Mr. Brown. Again, back to the questions before, what is your 
thinking as the person in charge? 

Mr. O'Connor. Sir, I think there are multiple costs involved; and 
if there are legitimate costs that we should reimburse, then the 
University's position is that it will reimburse. 

Mr. Brown. What are legitimate costs? Give me some examples 
of what you would consider legitimate costs that you should reim- 
burse for. 

Mr. O'Connor. I don't have one off the top of my head, Congress- 
man. 

Mr. Brown. Give me costs that you shouldn't reimburse for. 

Mr. O'Connor. That we should not reimburse 



166 

Mr. Brown. The government for. Since you can't come up with 
any costs that you should reimburse us for, are there any that you 
shouldn't reimburse us for? 

Mr. O'Connor. There may be costs that the Federal Government 
should be reimbursed for, and as I said, the University will do that, 
but I don't have a compilation of those costs in front of me right 
now, sir. I will be happy to attempt to provide the committee with 
such. 

Mr. Brown. What is your role — what is your view. Dr. O'Connor, 
about the role of the University in overseeing NSABP with respect 
to the follow-up of the fraudulent activities of St. Luc in Montreal? 

Mr. O'Connor. I think it is a very important role, Congressman, 
and I think we intend to take it very seriously, as Dr. Herberman 
has pointed out. 

Mr. Brown. Did the University take any steps to ensure that Dr. 
Fisher and his colleagues publish any reanalysis of the data in a 
timely manner? 

Mr. O'Connor. We have requested that Dr. Fisher publish such 
materials. 

Mr. Brown. Prior to the public revelation of the fraud, did you 
ever take any steps to ensure that he published a reanalysis of the 
data? 

Mr. O'Connor. Do you mind if I ask Dr. Detre some information 
on that? 

Mr. Detre. Congressman Brown, the University, sometime I be- 
lieve in 1990, created a new office to investigate scientific integrity 
precisely because Mr. Dingell's committee has been critical of the 
quality of examinations we have done at the University. In that 
process — if you don't mind, I would like to describe the structure 
to you just in a sentence or two. 

In this new office — it reports directly to the chancellor of the uni- 
versity via university counsel — the provost and the senior vice 
chancellor for health sciences are notified only when action has to 
be taken. 

In the case of Dr. Poisson in Montreal, I have learned just a few 
weeks ago that our offiice notified the dean of the school of medicine 
that ORI has completed its investigation and instructed NSABP to 
do an analysis and publish the information. I was not notified 
about that. Congressman Brown, and maybe this degree of inde- 
pendence that the office currently has, which was set up for the 
protection of the university, has gone beyond its usefulness. 

Mr. Brown. Dr. Detre, in your prepared statement, you spoke 
about the culture of deference. 

Mr. Detre. Yes, sir. 

Mr. Brown. Culture of deference in academic settings toward a 
university senior scientist, something that obviously — I would 
think is not unique to the University of Pittsburgh certainly. Did 
that culture of deference at Pittsburgh toward Dr. Fisher, did that 
contribute to the university's failure of oversight? 

Mr. Detre. I believe it contributes to any university's failure be- 
cause we don't have a mechanism in place to truly supervise senior 
faculty. Our expectation. Congressman Brown, is that if there are 
problems, they will be reported by the senior faculty member to the 



167 

chairman of the department and, through him, to the dean; but 
clearly, this mechanism alone is insufficient. 

Mr. Brown. You told the subcommittee staff in an informal sur- 
vey that 90 percent of Pittsburgh's academic peers treat senior sci- 
entists in a sort of similar hands-off, detached fashion in a sense. 
Is this, in fact, a problem throughout the University? 

Mr. Detre. Congressman Brown, I think what I said to the dis- 
tinguished staff of Mr. Dingell's committee is that every university 
has the same tradition of deference, that we are no exception to. 

Mr. Brown. In light of what has happened at your very fine uni- 
versity, what steps should the university take, or what steps do 
you plan to take, to grapple — to deal with this whole cultural def- 
erence to assure accountability and to be responsive to the public 
trust when Federal dollars or any government dollars are involved? 

Mr. Detre. Congressman Brown, as I earlier said to a question 
posed by Congressman Schaefer, I believe that an external quality 
assurance and auditing program is absolutely essential to guaran- 
tee the proper administration of these matters. 

Mr. Brown. Is the University of Pittsburgh going to lead the 
way? 

This culture of deference that you have labeled clearly can cause 
potential problems in university settings across the country. Do you 
see any role for the University of Pittsburgh to serve to lead the 
way in beginning to undo some of that? 

Mr. Detre. We intend to be the leader in this, though I fully rec- 
ognize. Congressman Brown, that it will not necessarily win a pop- 
ularity contest. 

Mr. Brown. Talk to me a little about what "the culture of def- 
erence" means. I understand what it means exactly in this NSABP 
case, where there wasn't a challenge to things he was saying, but 
does it include things like these lavish conferences in Hilton Head 
and first class airfare and even — 

Mr. Detre. I don't think that has anything to do with it. 

You know, as distasteful as it may appear to you, I don't believe 
that is the real problem. The real problem is that when deficiencies 
are identified — and I would like to remind you. Congressman, that 
actually it was NSABP that discovered the cheating at Montreal, 
not another agency; so the difficulties we find ourselves in — that 
we have no way of monitoring whether all necessary corrective ac- 
tions have been taken or not. This is our major deficiency. 

Mr. Brown. Tell me some specific examples of some of the nega- 
tive outgrowths of this cultural deference, how it is manifested in 
a way that does not add to public knowledge, might potentially 
abuse the public trust, is more of a negative than a positive. 

Mr. Detre. Congressman Brown, do you want to see an even 
more vivid example than what we are witnessing now, the reason 
why we are here? 

Mr. Brown. Sure. 

Mr. Detre. I think it is a perfect example of the difficulties uni- 
versities find themselves in, because we do not have a university- 
based but independent quality assurance and audit program which 
would help us to identify problems and correct them. 

Mr. Brown. Why not? 



168 

Mr. Detre. It has not been part of our culture, Congressman. 
The society learns gradually from its mistakes. 

I remember that 15 years ago neither Congress nor universities 
had any conflict-of-interest policies. We now have them. I think 
this is a gradual growing process. We learn from our mistakes, and 
we try to rectify them. 

Mr. Brown. Dr. O'Connor, I assume you are going to accelerate 
some of that learning process? 

Mr. O'Connor. I certainly hope to, Congressman Brown. 

Mr. Brown. Thank you, gentlemen, for testifying, for being with 
us today. We will take a short, 15-minute recess or so, while the 
chairman and the rest of us vote. Then we will call Dr. Fisher im- 
mediately upon returning. 

[Brief recess.] 

Mr. Dingell. The Chair advises the next panel is composed of 
Dr. Bernard Fisher, M.D., former chairman. National Surgical Ad- 
juvant Breast and Bowel Project. 

Dr. Fisher, welcome to the committee. 

The Chair advises that all witnesses are sworn before the com- 
mittee. Do you have any objection to being sworn in? 

Mr. Fisher. No, sir. 

Mr. Dingell. Very well. The Chair advises that you are, since 
you will be testifying under oath, entitled to be advised by counsel. 
Do you so desire? 

Mr. Fisher. Yes, sir. 

Mr. Dingell. Your counsel is seated next to you. You are? 

Mr. Onek. Mr. Joseph Onek, Mr. Chairman. 

Mr. Dingell. How do you do, sir. The Chair will note for the 
record that you will then be advising Dr. Fisher. The Chair advises 
that copies of the rules of the subcommittee, the committee, and 
the House are there at the witness table to assist you as you testify 
before the committee. 

If you have, then, Doctor, no objection to testifying under oath, 
if you will please rise and raise your right hand. 

[Witness sworn.] 

Mr. Dingell. You may consider yourself then under oath, Doc- 
tor, and the Chair will recognize you for such opening statement 
as you choose to give. 

TESTIMONY OF BERNARD FISHER, FORMER CHAIRMAN, NA- 
TIONAL SURGICAL ADJUVANT BREAST AND BOWEL 
PROJECT, UNIVERSITY OF PITTSBURGH, ACCOMPANIED BY 
JOSEPH ONEK, COUNSEL 

Mr. Fisher. Thank you, sir. Mr. Chairman, members of the sub- 
committee, I am grateful to have the opportunity to be here today. 
I am Dr. Bernard Fisher, distinguished service professor of surgery 
at the University of Pittsburgh. Over the years, I have been a 
member of the President's Cancer Panel, the Board of Scientific 
Counselors, and the National Cancer Advisory Board of the Na- 
tional Cancer Institute. 

I have passionately devoted more than 35 years to the study and 
treatment of breast cancer to the exclusion of almost everything 
else in my life. My studies involving nearly 50,000 women and 



169 

5,000 health professionals at 500 institutions in North America 
have, I believe, revolutionized the treatment of that dread disease. 

As a result of the research by the NSABP, which I chaired for 
27 years, women with breast cancer now have a choice to save their 
breasts rather than to undergo a disfiguring radical mastectomy. 
Our work has shown that chemotherapy and tamoxifen after sur- 
gery increases survival. 

We recently began the study to determine whether breast cancer 
can be prevented in women at high risk. If the findings from that 
trial indicate that the risk of breast cancer can be reduced, the 
problem of breast cancer would be significantly diminished. 

The events arising out of the data falsification in Canada have 
been tragic for me, my colleagues, our families, and for all women 
in this country. Women with breast cancer began to doubt the ther- 
apy they had received. Those without breast cancer lost confidence 
in the system that they someday might need to rely upon. 

Mr. Chairman, I can't emphatically enough state that women 
must not become the victims of these events. In my statement, I 
will address point by point specific matters that have arisen during 
the past few months. 

First, I sincerely share the subcommittee's concern regarding 
fraud in science and believe that any deviation from scientific in- 
tegrity is not acceptable. My whole scientific life has been based on 
that credo. I deeply regret that there was data falsification by a 
physician at one of the hospitals participating in the NSABP. 

Second, despite the data falsification, women can and must feel 
secure that lumpectomy following radiation therapy is as effective 
a treatment as removal of the breast. Our studies, as well as those 
of others, have consistently supported our findings. The NCI, upon 
recent review of our results, reached the same conclusion. 

Third, there never was any intent to hide information regarding 
the discovery of falsified data at St. Luc Hospital in Montreal, and 
I emphasize that. There never was any intent by me or my associ- 
ates to hide any information regarding the discovery of that infor- 
mation. 

On November 14, 1990, I was notified by Dr. Carol Redmond, Di- 
rector of the NSABP Biostatistical Center, that data discrepancies 
found at St. Luc Hospital were being investigated. At that point, 
no fraud had been identified. 

In February 1991, following further investigation by the 
Biostatistical Center, it was found that data falsification had oc- 
curred. We immediately notified the NCI and the investigator. Dr. 
Roger Poisson, was suspended. 

We believe that the detection and reporting of the St. Luc Hos- 
pital fraud was carried out according to the 1988 clinical trials co- 
operative group program guidelines which state that, "In more seri- 
ous cases, it is the responsibility of each group to define the gravity 
and degree of potential problems and to be consistent and fair in 
the actions and sanctions it applies when significant problems are 
uncovered." The time spent by the NSABP investigation was used 
to ensure that a fair action was taken. 

After we alerted NIH, an official investigation was begun by the 
Office of Scientific Integrity, OSI. We were embargoed from further 
discussions of the matter. On several occasions, the data from 



170 

NSABP studies in which Dr. Poisson participated were reanalyzed, 
eliminating his patients. 

In March of 1992, the findings were shared with members of the 
NCI, OSI, and NIH. These findings indicated that when all data 
from St. Luc Hospital were removed from the reanalyses, the out- 
comes and conclusions of all of our previously published studies 
were unchanged. We had already notified the NSABP Executive 
Committee of the Poisson affair and of the results of our reanalyses 
in February of 1992. 

The OSI thorough investigation found falsification in 99 breast 
cancer patients in 22 studies. These falsifications represented 0.3 
percent of the 33,885 women in our studies; 98 of the 99 alterations 
were related to patient entry criteria. In randomized studies, such 
types of falsifications are unlikely to influence patient outcome. 

Let me emphasize this firmly, if our reanalyses had produced 
evidence that the conclusions of our studies were affected by the 
data alterations, we would have reported our findings immediately. 
Neither we, the NCI, the NIH, nor the OSI perceived that the 
Poisson falsifications had resulted in a public health problem. If 
the NCI had considered this to be the case, they could have issued 
a clinical alert. 

Our plan was to present our findings to the scientific community 
in a peer-reviewed publication and to prepare a comprehensive 
technical report. We didn't realize that the failure to publish our 
findings immediately would be misinterpreted by the public as an 
indication that we were concealing information. Such a perception 
resulted in the unjustified concern that women with breast cancer 
were receiving inappropriate therapy. 

We should have been more sensitive to this possibility and we 
should have published our reanalyses more promptly, and I truly 
apologize for that delay. 

For 30 years, the goal of the NSABP was to provide better infor- 
mation to patients and their physicians. The suggestion that we 
suppressed information is painful. We never attempted to hide any 
information. There could have been no conceivable reason for us to 
do so. 

Fourth, concern has been raised regarding inclusion of data from 
St. Luc Hospital in NSABP reports published after February 1991, 
when the Poisson fraud was found. There are honest differences of 
opinion among statisticians regarding the handling of falsified data 
in large clinical trials. There are significant scientific reasons not 
to exclude all such data. 

Our statisticians considered that it was not appropriate to ex- 
clude all data from those patients who had a diagnosis of breast 
cancer and who had been randomized, treated, and followed appro- 
priately. Excluding all patients would have prevented identification 
of toxicities and other adverse events. 

During the OSI embargo, we could not exclude data without dis- 
cussing the findings with journals; and after the embargo was lift- 
ed, we complied with the NCI request of January 1993 that no data 
from St. Luc be included in publications containing new data. 

A fifth concern relates to the reporting of deaths from 
endometrial cancer in breast cancer patients treated with 
tamoxifen. As was mentioned previously, there are more than 4 



171 

million women using that drug. Clinical trials have demonstrated 
that the benefit from tamoxifen is about 20 times greater than the 
risk of getting endometrial cancer. Consequently, women with 
breast cancer should continue to take tamoxifen, although some un- 
desirable effects may occur. 

In our studies, we found that of 2,693 breast cancer patients 
being treated with tamoxifen, 23 developed endometrial cancer and 
4 deaths occurred. There is no substance to the suggestion that we 
withheld information concerning these four deaths. 

It is important to understand the difficulties involved in deter- 
mining the cause of death in breast cancer patients with 
endometrial cancer. When a death occurs in a woman who had both 
breast cancer and endometrial cancer, it is extremely difficult to be 
sure which cancer caused the death. It cannot be assumed that the 
death occurred as a result of the endometrial cancer; she may have 
died of breast cancer. Only by continuing medical review and 
rereview and by obtaining difficult-to-get information can we be 
sure which cancer caused the death. Such medical detective work 
can take a long time. 

There is often delay in obtaining information about deaths be- 
cause patients move, go to different hospitals, change physicians or 
because families fail to notify physicians about a death. These de- 
layed responses can result in a delay in confirming the cause of 
death. Death certificates can be ambiguous or inaccurate and not 
readily obtainable. Autopsies are infrequently performed and often 
fail to aid in determining the cause of death. 

In the fall of 1993, we confirmed that there had been deaths from 
endometrial cancer in tamoxifen patients and reported this fact to 
the NSABP, the NCI, and the drug manufacturer. In retrospect, it 
might have been possible to collect and report information about 
these deaths sooner, but there was never any intent to withhold 
any information. 

Sixth, concerns have been raised about the NSABP audit proce- 
dures. The NSABP verifies and evaluates the quality of data sub- 
mitted by a quality assurance program which has two parts. One 
is the on-site audit program in which visits to participating institu- 
tions are made at least once during a 3-year period by NSABP per- 
sonnel. Selected samples of patient records are examined. 

In the second part of the program, extensive source records from 
all patients are received and examined at the NSABP head- 
quarters. In 1991, the reviewers of our grant application to the NCI 
termed our procedures "exemplary." 

Since 1991, there have been significant changes in the workload 
of the NSABP. The initiation of the breast cancer prevention trial 
has resulted in an increase in the number of patients being fol- 
lowed in NSABP studies from 25,000 in 1991 to 41,000 in 1993; 
and the number of data forms processed expanded from 225,000 in 
1991 to 413,000 in 1993. 

In retrospect, the administrative infrastructure of the NSABP 
did not keep pace with this tremendous growth. There have been 
some delays in our auditing and reporting functions. 

I accept my share of the responsibility for these administrative 
deficiencies that occurred; I could have been more aggressive in 
seeking funding for additional administrative personnel. In retro- 



172 

spect, I should have brought on more executive administrative peo- 
ple to help manage the program, an executive director. Perhaps my 
passionate attention to the science overshadowed my administra- 
tive insight, and this was a mistake. I should have been firmer 
with the personnel responsible for the audit program. 

I believe that the administrative deficiencies of the NSABP could 
be remedied quickly. For this reason, I am troubled by the pro- 
longed suspension by the NCI of all NSABP clinical trials in 
progress and the postponement of the development and start of 
new studies. These events could affect the lives of thousands of 
breast cancer patients in years to come. Clinical trials must be re- 
stored at once. 

Women and their doctors must realize that large multicenter 
randomized trials are the best way to obtain information for mak- 
ing treatment decisions. The randomization of patients in trials 
eliminates the chance of obtaining biased results. 

Clinical trials also provide — and I emphasize this — higher stand- 
ards of patient care. If clinical trials are eliminated, we have no 
good alternatives. We cannot return to the system where treatment 
is based on physicians' intuition. 

Finally, I thank the NCI for the support which made my life's 
work possible. For 25 years, NSABP's relationship with officials of 
the NCI was based upon mutual respect and cooperation. Members 
of the NCI have been involved in every aspect of our efforts. They 
have continuously been aware of and have participated in our suc- 
cesses and were intimately involved with the decision-making proc- 
ess of our group. This was truly a cooperative agreement in name 
and in fact. 

It is necessary for the good of the women in this country that we 
now look to the future. Within the last 6 months, I have formulated 
a strategic plan for a new generation of clinical trials, which I 
hoped would be my legacy, which could significantly alter the fu- 
ture treatment of breast cancer patients. 

One such study evaluates the use of preoperative therapy, and 
if the results indicate, primary surgical treatment for breast cancer 
will become obsolete. Information from another study could more 
clearly define a patient's risk for developing a treatment failure 
and consequently identify which patients should or should not re- 
ceive a particular therapy. Other studies by us are evaluating new 
and exciting therapeutic agents that can be better than those cur- 
rently used. 

For progress to be made in breast cancer treatment, the public's 
faith in clinical trials must not be diminished by these current 
events or any other. 

In conclusion, I emphasize that any deviation from scientific in- 
tegrity is unacceptable. While fraudulent data were submitted by 
a contributing investigator, the NSABP studies themselves are not 
fraudulent. There was never any intent to hide information or de- 
ceive anybody. During the course of all our activities, NCI and 
Zeneca were intimately involved. 

System errors can and did occur and must be repaired. I believe, 
as was heard, improved technology will help. Above all, we must 
not allow these recent events to deflect us from our ultimate goal — 
the prevention and cure of breast cancer in women. 



173 

Thank you, sir. 

Mr. DiNGELL. Thank you, Dr. Fisher. The Chair is going to recog- 
nize himself first. 

Doctor, could you tell the subcommittee the importance of an 
audit program to a clinical trial such as the one you headed at 
NSABP, please? 

Mr. Fisher. Yes, sir. The purpose of an audit program is to ver- 
ify the quality of information that is being obtained and is being 
submitted to the NSABP Headquarters. 

Mr. DiNGELL. Now, Dr. Fisher, during the course of the trials 
conducted at NSABP, your staff conducted literally hundreds of 
these audits; is that correct? 

Mr. Fisher. Yes, sir. There were 587 audits conducted in four cy- 
cles. 

Mr. DiNGELL. Now, Doctor, tell me who is Dr. Wickersham or 
Wickerham? 

Mr. Fisher. Dr. Wickerham is the Deputy Director of the Oper- 
ations Office. 

Mr. DiNGELL. He is the chief medical person in the study, is that 
right, and the auditors reported to him? 

Mr. Fisher. The audit program has two parts. The audit pro- 
gram has a part which is conducted by the Biostatistical Center. 
The Biostatistical Center is the one responsible for scheduling au- 
dits, sending out people to audit the program, to bring and to ex- 
amine the things that they do at the site, and to bring back to the 
Headquarters source information, reports of x rays, all kinds of 
things, laboratory information, information about treatment, and so 
on; and then at the Headquarters, by the Biostatistical Center, this 
is all reviewed to make sure that what was sent in originally is 
that which exists in the source document. 

Mr. DiNGELL. Dr. Wickerham told the subcommittee staff that he 
routinely gave you copies of the audit reports at the NSABP sites; 
is that correct? 

Mr. Fisher. I received audit reports. I cannot say with certainty 
whether these were routine or all audit reports, but I did receive 
audit reports. 

Mr. DiNGELL. Now, Doctor, the subcommittee staff has been re- 
viewing the audit reports for the last month or so, and it has found 
that in addition to the St. Luc problems relative to fraud, your 
auditors uncovered a number of similar problems at sites through- 
out the NSABP. 

For example, a number of locations were found to be violating 
the eligibility criteria. At one site, % of the patients enrolled did 
not meet the eligibility criteria. Can you tell us about what you 
knew about these matters regarding the depth and the breadth of 
eligibility problems identified by your auditors? 

Mr. Fisher. Mr. Chairman, we recognized that there were ad- 
ministrative deficiencies in the audit program, but with respect to 
any particular institution, we have not had a chance to review the 
subcommittee's investigations or reports or any other recent analy- 
ses, and we would certainly like to have the opportunity to do so 
and provide the information to you. There may be some misunder- 
standings regarding clarification of ineligibility and other terms 
which exist. 



174 

Mr. DiNGELL. I am not going to refer to the specific sites unless 
it appears to be desirable so to do, but these were audits which 
were performed over the years, clear back into the 1990's, the early 
1990's, and on back even beyond that into the 1980's. And that was 
at a time when you were in charge of the program, and didn't they 
give you some awareness of the fact that there were problems with 
either fraud or slovenly work? 

Mr. Fisher. Well, it was the audit process which did uncover the 
St. Luc; it was the part of the quality assurance program which un- 
covered the St. Luc falsifications. We found those falsifications 
through our efforts, and reported them, and that is the only fal- 
sification that has been 

Mr. DiNGELL. Of course. Doctor, these went to eligibility and 
showed that there were some fairly significant eligibility questions. 
For example, one site had three-quarters of the participants ineli- 
gible. 

Mr. Fisher. I am certainly unaware of that, sir. I really am not 
aware of it. 

Mr. DiNGELL. But it was in audit reports that came to you, 
though. 

Mr. Fisher. I don't, I really don't remember seeing that report 
at all. 

Mr. DiNGELL. Well, here they are, and I am just going to lay 
them out. South Nasau Hospital, Rush Presbyterian Hospital, St. 
Joseph Hospital in Lancaster, and in the University of Pittsburgh. 
Now, the University of California Davis had % of the participants 
in the study ineligible. 

Mr. Fisher. I have never seen that. 

Mr. DiNGELL. These were audits, these were your audits in your 
project. 

Mr. Fisher. May I make a comment about eligibility? 

Mr. DiNGELL. Sure. 

Mr. Fisher. Eligibility has been — the term "eligibility" has been 
used here and elsewhere. Eligibility is not falsification. 

Mr. DiNGELL. We are not — I am not making the allegation that 
eligibility is a fraud, or fraud or similar question. It is, however, 
a matter which goes to the very reality and scientific adequacy of 
the test, because if you are testing people or reporting on people 
who don't meet your eligibility requirements, it tends to skew your 
results. And what I am trying to find out is, all of these matters 
were essentially found in audits, but we are not able to address 
them here today because of your lack of familiarity with them. 

Mr. Fisher. I certainly am not aware of any institution where 
there were % of the patients ineligible. I should like to have more 
information about that than I have. 

Mr. DiNGELL. Well, we will give it to you. It is, however, I would 
observe. Doctor, in your own audit reports. 

Now, here are other examples. Auditors turned up instances 
where patients were randomized twice. What is the result of ran- 
domizing a patient twice in a study of this kind? What does it do 
to the statistical validity of the study? 

Mr. Fisher. I can't answer that question. 

Mr. DiNGELL. Do you know how much double randomization was 
occurring and what the practical effect of it was? 



175 

Mr. Fisher. Sir, I don't know the number of double 
randomizations, but I must think they were extremely few and far 
between. This has not been brought to my attention as being a 
problem. 

Mr. DiNGELL. Well, your auditors also identified a number of in- 
formed consent problems throughout the site — throughout the dif- 
ferent sites. For example, no documentation of informed consent 
obtained up to 2 years — obtained up to 2 years post randomization 
and so forth. Now, what — first of all, does this constitute a problem 
in terms of lack of adequate informed consent by the participants, 
and if so, what was done? 

Mr. Fisher. Let me say emphatically that informed consent is a 
very important part of what we are doing. It always has been. 

Mr. DiNGELL. It is really an ethical question. 

Mr. Fisher. Absolutely. There is no question about that. And as 
far as I am concerned, that is something that there can be no ex- 
cuses for. There are certain situations where in one of the studies, 
for example, wherein formed consent may have been obtained after 
an operation was done that is in a particular study of lumpectomy 
where prerandomization was used, but in reference to your com- 
ments about informed consent, I have seen in some of the audit re- 
ports that there was this kind of situation, but I have not been fa- 
miliar with it as any kind of a serious problem. 

Mr. DiNGELL. Well, your auditors found a number of sites that 
were not maintaining drug laws. Can you tell us the importance of 
the drug laws — rather the drug logs and what happened at the lo- 
cations that were not properly maintaining these logs? 

Mr. Fisher. Drug logs are also something that is looked for at 
the site routinely to make sure that the drugs that are given to the 
investigators are used for the patients that they are supposed to 
be used on. 

Mr. DiNGELL. Well, but here you have another instance where 
the drug logs were not maintained. Is that important, or is that not 
important? 

Mr. Fisher. As far as I am concerned, it is important. 

Mr. DiNGELL. Now, a number of other locations had serious prob- 
lems with missing data. Were you aware of this? 

Mr. Fisher. I have heard about that, particularly in certain insti- 
tutions, and it depends again on how long ago the data was col- 
lected. For example, at one of the — in New Orleans, for example, 
where data was collected in the 1970's at a large city hospital, 
some of that data now in 1994 may be hard to obtain. I cannot 
know more than that about it, however. 

Mr. DiNGELL. Well, if the failure to maintain proper logs and to 
have the data properly assembled is recent, is that a more serious 
problem? 

Mr. Fisher. These things are all problems. There is no question 
about that. And I would certainly like to address them after I knew 
more about the degree of these problems. For example, I indicated 
to you that the NSABP had conducted 587 audits since 1982, and 
of those 587 audits, there were only — there were major problems 
identified in 5.8 percent, and some of those problems deal with the 
things that you are talking about. 



176 

Mr. DiNGELL. You mean there were problems identified in 5.8 
percent of the audits? 

Mr. Fisher. Which were considered to be serious enough to sus- 
pend these investigators. 

Mr. DiNGELL. Well, these involved some 30 institutions. Are you 
able to tell me that these were trivial matters or that they were 
not important matters? 

Mr. Fisher. No, sir. I in no way make any indication that these 
are trivial. We do the audits to determine these things. Otherwise, 
there would be no point in doing the audit program. 

Mr. DiNGELL. Well, I think you raise an important question. 
What did you do with the audits? The audits come in, they said, 
for example, there is a problem with informed consent. They say 
that a number of sites were not maintaining drug logs. They say 
that there was a serious problem with missing data. Did you in- 
quire into these problems? 

Mr. Fisher. The usual process would be that when this happens, 
the medical auditor, medical reviewer, will write a report, and that 
report will indicate to the investigator what the problems are and 
we would expect them to implement a plan of action to tell us what 
they are going to do to correct these problems. And then all this — 
this report also goes to a quality assurance committee which we 
have. 

We have a standing committee where all of these reports go to 
their review and their suggestions as to what kind of punitive ac- 
tion should be taken. The investigator is notified about these; he 
is — he or she is supposed to provide the NSABP with a plan of ac- 
tion that will be acceptable, and if it is acceptable, then the deci- 
sion is made to give them a chance to show that they have cor- 
rected themselves and resume accrual. If it is not acceptable, ac- 
crual continues to be suspended. 

Mr. DiNGELL. Well, that is all very good. Were you ever made 
aware of the fact that there were problems with informed consent? 
Were you ever made aware of the fact that there were problems 
with sites not maintaining drug logs? Were you ever made aware 
of the fact that a number of locations had problems with missing 
data? 

Mr. Fisher. As I said, I have seen these reports sent to me by 
Dr. Wickerham. At this moment I don't know what was in the re- 
ports, how many of them were related to informed consent, how 
many of them were related to drug log problems. I am unable to 
answer that, sir. 

Mr. DiNGELL. Well, you are able to tell us what was done about 
these matters. What was done about the informed consent ques- 
tion? What was done about the questions on the number of sites 
that were not maintaining drug logs? What was done about the 
sites where there were problems with missing data? What action 
did you take? 

Mr. Fisher. Well, we sent the reports back. One of the purposes 
of an audit program, aside from what I mentioned, is also that it 
is supposed to be and is a program, an interactive program where 
investigators are informed about their deficiencies and are edu- 
cated against doing — repeating this. 



177 

Mr. DiNGELL. What did you do about these matters to correct the 
situation, either in general or in any particular case? Did you do 
anything? Can you tell us any one thing you did on any one of 
these audits which came to your attention? Did you do anything 
about any of the audits which came to your attention with regard 
to either the questions of informed consent, questions of inadequate 
drug logs, or failure to maintain drug logs, and also locations that 
had problems with missing data? Can you tell us anything you did 
about any one of the audits on any one of these points? 

Mr. Fisher. I certainly — my main issue here would have been to 
order the personnel who were responsible for this program to carry 
out what they were supposed to do. 

Mr. DiNGELL. Well, what were you doing while these people were 
supposed to do these things? 

Mr. Fisher. I think what we were doing, sir, is that the NSABP 
did have continuing, ongoing workshops, meetings for data man- 
agers, for all kinds of people to educate them into — to try to get 
them to prevent this kind of practice. This was what was being 
done. 

Mr. DiNGELL. What was your job at NSABP? 

Mr. Fisher. My main role in the NSABP, and I take — but let me 
emphasize that I take full share of responsibility — full responsibil- 
ity for the administrative errors which took place under my term, 
but I was, as I mentioned in my introductory statement, my main 
interest was to be — to try to do the best science that would be pos- 
sible, which would affect the most women in the world with breast 
cancer. I was on top of the development of the scientific program, 
the implementation of the scientific program, the pooling together 
of the information and the publication of the science, and these 
were the chief efforts that I conducted. 

Mr. DiNGELL. Well, let's look. What was your job at the NSABP? 

Mr. Fisher. As I say 

Mr. DiNGELL. You were the head of the whole operation, were 
you not? 

Mr. Fisher. Yes, sir. 

Mr. DiNGELL. And you got these audit reports? 

Mr. Fisher. I take responsibility for the operation; I wasn't the 
head of all of the pieces that were under me. The biostatistical cen- 
ter, the data center which is a major part of the NSABP; it actually 
receives more of the funding than does the operations center. 

Mr. DiNGELL. Who was responsible for management and admin- 
istration? 

Mr. Fisher. The management and administration was delegated 
in certain areas. The fiscal administration 

Mr. DiNGELL. To who? 

Mr. Fisher [continuing]. — Was given to a fiscal officer who had 
been with me for 20 years who had a staff of people that conduct 
all of the fiscal affairs in concert with the University of Pittsburgh 
fiscal office. All of that was integrated. 

Mr. DiNGELL. What did this individual do with the audits? Did 
he do anything with them? 

Mr. Fisher. At one time Ms. Dash, who was the — was the fiscal 
officer, her job was to — all of the audit reports came to her before 



178 

they were finally sent out, and it was her job to see that they did 
get distributed. 

Mr. DiNGELL. Did this person report to you? 

Mr, Fisher. She always reported to me as 

Mr. DiNGELL. She was responsible to you, was she not? 

Mr. Fisher. Yes, sir. 

Mr. DiNGELL. And you were responsible for her; is that right? 

Mr. Fisher. Yes, sir. 

Mr. DiNGELL. Now, the chief auditor, Marjorie McLaughlin, told 
the subcommittee staff there were a number of chronically prob- 
lematic sites, such as Tulane and LSU. Were there any special ef- 
forts taken to try to deal with the sites that seemed to come up 
year after year to exhibit an inability to follow the protocols of the 
study? 

Mr. Fisher. Yes, sir. In talking with the principal investigator on 
many occasions by myself and by our staff, there was a Catch-22 
in that situation. 

Mr. DiNGELL. What was the Catch-22, Doctor? 

Mr. Fisher. Well, the Catch-22 was simply this, and this has 
been a part that we had experienced over many years. The institu- 
tions that put on the larger numbers of patients who did put on 
large numbers of patients, if they had problems and if those insti- 
tutions were to be eliminated from the NSABP, the backlog, the 
large numbers of patients that needed to be followed up still re- 
mained there to be followed up. And so we did keep those people, 
when we suspended them, we kept them on for at least follow up, 
until 

Mr. DiNGELL. Well, I have no quarrel with follow up; I will con- 
cede that this is very important. But that would be a part of the 
research protocols in any event, that they had to follow up; would 
it not? If only to assure the health and the safety and the well- 
being of the patient. 

Mr. Fisher. Absolutely. 

Mr. DiNGELL. All right. So they had that responsibility, whether 
they remained in the test program or not. But was there ever any 
disciplinary action taken against Tulane or LSU or any of the oth- 
ers which were either chronic problem areas or which failed to deal 
properly with the problem of informed consent, or which failed to 
keep proper drug logs, or which had serious problems with missing 
data? 

Now, we have established that you had all of these difficulties, 
and I am trying to find out what one thing that you did with re- 
gard to any of the sites which were deficient, some of which, as we 
have indicated, were chronically problem sites and others of which 
had significant failures. 

Now, what one thing did you do or what one thing did anybody 
else do about these sites which were problems? 

Mr. Fisher. We suspended their accrual and gave them 

Mr. DiNGELL. When did you do that? How many of them did you 
do that to? 

Mr. Fisher. How many of them did we do that to? In the last 
cycle 4 in which we did 154 audits — 129 audits; I think there were 
6 that were suspended. 



179 

Mr. DiNGELL. But they were suspended, and for how long were 
they suspended? 

Mr. Fisher. As I recall, for varying periods of time, depending 
upon what the problem was, how rapidly they could get it cor- 
rected. I can't — I don't 

Mr. DiNGELL. Who were the six that were suspended? 

Mr. Fisher. I don't have that at my fingertips. 

Mr. DiNGELL. Were Tulane or LSU suspended? 

Mr. Fisher. Not in the last cycle, no. Earlier, but not in the last 
cycle. 

Mr. DiNGELL. But your auditors tell us they were chronic prob- 
lems. Now, what did suspension mean? Did suspension mean that 
they were removed from the test, they were removed from the test 
until they had corrected the problems? 

Mr. Fisher. They were unable to put any more patients on to the 
studies until those problems were corrected. We hoped to keep 
them there, to keep the patients followed that were there, and that 
was it. 

Mr. DiNGELL. But they continued to participate with the patients 
that they had on in the program; is that right? 

Mr. Fisher. They continued, they continued, but didn't have any 
activity, other than what you say. 

Mr. DiNGELL. OK. So they are continuing to submit data. 

Now, have you suspended, let's say, institution X for bad data 
and missing data? You kept them on to continue with their efforts 
and to continue supplying missing data? 

Mr. Fisher. No. We certainly did not. We tried to provide them 
with 

Mr. DiNGELL. Well, you told us you had kept them on and you 
kept them on. You didn't allow them to enroll new patients, but 
you kept them on. So you kept them on and they continued the 
same practices. They continued the same practices with regard to 
failure to achieve proper informed consent, failure to maintain 
proper drug logs, and failure to provide adequate data. 

What I am trying to find out is what disciplines did you have, 
what did you do, how did you — how were you able to tell us when 
your own chief auditor tells us that there were a number of chron- 
ically problematic sites, and I am asking you what did you do about 
any one of these to see to it that they corrected their bad practices? 
What did you do about to see that they corrected their bad prac- 
tices? You suspended them, but they continued apparently accord- 
ing to your chief investigator, providing bad information, and en- 
gaging in other practices which, quite frankly, raise questions 
about the value and the integrity of the scientific process involved 
in this particular study. 

Mr. Fisher. Sir, I cannot answer these questions unless I know 
the specific cases and so on. I am sorry. We continue to try to edu- 
cate these people. 

Mr. DiNGELL. It is my impression that you were the man that 
ran this whole thing. 

Mr. Fisher. The — in terms of our involvement with these prin- 
cipal investigators, we did talk to them; we did try to educate their 
data managers; we did try to have affirmative kinds of things in 
that nature. 



180 

Mr. DiNGELL. But your auditors continued to find chronic prob- 
lems. 

Mr. Fisher. I don't know. 

Mr. DiNGELL. Didn't they tell you, we have a problem here, and 
didn't you in looking at the audit say, by golly, we saw them last 
week or last month or last year and we saw them the month before 
that and the year before that and the month before that and the 
year before that? Were you paying any heed at all to these matters, 
Doctor, or were you delegating it to somebody who was not report- 
ing to you, or were they reporting to you and you not paying heed 
to them? 

Apparently the auditors sent you reports with which you do little 
or nothing, and apparently the chief auditor tells us about a num- 
ber of chronic problems sites. And apparently the people whom you 
delegated to deal with the administrative end of the business didn't 
deal with the administrative end of the business in terms of bring- 
ing it to your attention, or if they did, you didn't pay heed to it be- 
cause the situation went on over the entirety of the years. What 
can you tell us about this? 

Mr. Fisher. I can tell you that I accept responsibility for any in- 
adequacies that took place, and I am very sorry for that, and 

Mr. DiNGELL. Here is another question. Let's take Memorial Can- 
cer Research Foundation, Dr. David Plotkin. They were identified 
by NSABP auditors in a report in 1990, and a transmittal letter 
with findings was to be shipped to Dr. Plotkin and to NCI. The 
signed letter and the envelope sat in NSABP files until recently 
and the audit report never left the NSABP. How did this happen? 

Mr. Fisher. I did not know about this until others knew about 
it. 

Mr. DiNGELL. Well, let's assume that it is true, and I believe that 
it is. You never notified the people about their failures. The audit 
report never left the NSABP. 

Mr. Fisher. I don't know, sir. It may not have or it may have. 

Mr. DiNGELL. I am going to — I have to go to the Floor; I am 
going to ask Mr. Brown to continue presiding. I will be back just 
as quickly as I can. 

Mr. Brown, would you take the chair please, sir. 

Mr. Brown [presiding]. I thank the chairman. The gentleman 
from Colorado, Mr. Schaefer. 

Mr. Schaefer. Thank you, Mr. Chairman. 

Dr. Fisher, after the media broke the story on the data falsifica- 
tion by Dr. Poisson, you purportedly reanalyzed the data and 
briefed ORI. Dr. Bivens told us in April that your presentation was 
oral and was not written; is that correct? 

Mr. Fisher. Presentation to the ORI, the NCI, NIH people was 
presented orally in 19 — March 1992 at their request. There had 
been many reanalyses done by the five statistical centers of the 
NSABP, Dr. Redmond and her associates. She did one almost im- 
mediately following the finding of the first — the proof of falsifica- 
tions. She did another one which was presented at this meeting 
that you are referring to. There were others that were done subse- 
quently and at one point material of a reanalysis was submitted to 
the NCI. So that there were repeated reanalyses. 



181 

And at the time of the one at the — one that was presented to the 
ORI, it is my recollection that there was total acceptance of this 
audit — of this reanalysis, and there were no comments either spo- 
ken or written to us regarding any concerns about the reanalysis. 

Mr. SCHAEFER. Well, so it was oral. And is this the common prac- 
tice not to do this in writing? 

Mr. Fisher. It is what we were asked to do, sir. 

Mr. ScHAEFER. Asked to do by 

Mr. Fisher. Asked to do by the ORI. 

Mr. Schaefer. By ORI? 

Mr. Fisher. Yes, sir. 

Mr. Schaefer. All right. Now, was this a complete reanalysis? 

Mr. Fisher. Yes, sir. Complete reanalysis of all protocols which 
Dr. Poisson had submitted patients to. 

Mr. Schaefer. Was this a running update, though, over a period 
of time where you kind of looked at new things all the time? 

Mr. Fisher. No, sir. No. It was a complete update at one time, 
one point in time. 

Mr. Schaefer. All right. At the hearing we had in April, we 
were told that it was not a complete reanalysis. 

Mr. Fisher. To the best of my knowledge, sir, and the best of my 
remembrance, it was 

Mr. Schaefer. And it has been done since then? 

Mr. Fisher. It has been done since then many times, yes. 

Mr. Schaefer. Yes, but since April? 

Mr. Fisher. It has been done since April, yes. I am terribly sorry 
that Dr. Redmond isn't here, because Dr. Redmond, who is the 
statistician who is in charge of all of this, would better address 
these issues than I can. 

Mr. Schaefer. Well, again, as the chairman spoke, you were in 
charge of this whole thing? 

Mr. Fisher. Absolutely. 

Mr. Schaefer. So therefore, you know, sooner or later the buck 
stops somewhere. 

Mr. Fisher. Right, sir. 

Mr. Schaefer. This Dr. Poisson, what is he doing these days? 

Mr. Fisher. I haven't the vaguest idea. 

Mr. Schaefer. After you were made aware of the fraud that had 
been perpetrate by Dr. Poisson, you stated that you immediately 
notified NCI. Tell me, did you ever notify the editors of the New 
England Journal of Medicine or any other publications about this 
so that some of your colleagues would know what was going on? 

Mr. Fisher. No, sir, because at that point in time we were em- 
bargoed by the ORI, for during their entire investigation, we were 
only permitted to present — to tell people about this under a need 
to know, and there was no — so to answer your question, no, we did 
not. 

Mr. Schaefer. Now, you say you were embargoed, but you still 
provided this to NCI, this information? 

Mr. Fisher. I am sorry. 

Mr. Schaefer. You say you were embargoed, but you still — you 
provided the information on the fraudulent findings to the NCI? 

Mr. Fisher. Well, that was certainly a need to know. 

Mr. Schaefer. But that was as far as it went, then? 



182 

Mr. Fisher. Yes, sir. We did get one — we requested from the ORI 
that we be permitted to report this to our NSABP executive com- 
mittee so that they knew what was — that this was an affair that 
was going on. 

Mr. SCHAEFER. What was the answer? 

Mr. Fisher. And we did give it to them, to the executive commit- 
tee. In February of 1992, we presented the fact that this was going 
on and we also presented them with one — with the reanalysis to in- 
dicate to them that taking all of the data out in all of these 
reanalyses failed to influence the outcome — either the outcomes or 
conclusions of any of our studies that had been previously reported. 

Mr. Schaefer. So therefore, now the embargo is lifted? 

Mr. Fisher. Yes, sir. 

Mr. Schaefer. OK. Then did you inform the New England Jour- 
nal of Medicine or any other publication of this, and if not, why 
not? 

Mr. Fisher. I guess the most direct answer to your question is 
that why we did not immediately or otherwise inform them is be- 
cause we really didn't know that this was necessary to do. 

Mr. Schaefer. Even though you knew there was fraud there? 

Mr. Fisher. The point is that doing all of these reanalyses did 
not in any way alter the outcomes or conclusions of our studies. 
There was no alteration whatsoever, and so therefore, that is 

Mr. Schaefer. All right. Let me get this straight, then. So, no, 
you did not inform any medical journals of this falsified data? 

Mr. Fisher. I am sorry. 

Mr. Schaefer. You did not inform any medical journals of this 
data after the embargo was lifted? 

Mr. Fisher. At that time, right. 

Mr. Schaefer. You did not. 

Mr. Fisher. Right. 

Mr. Schaefer. I have a little bit of trouble figuring out why you 
wouldn't when we have all of these women throughout the country 
who are looking at these procedures or possible procedures trying 
to figure out which way they should go and we have doctors all 
over the place that would, I would think, dearly love to have this 
information. 

Mr. Fisher. Well, as I have said, there was no evidence in any 
way that these data had been changed, that the outcomes had been 
changed, and therefore, we did not really perceive this as a major 
problem, public health problem or otherwise, and again, as I said 
in my remarks, had I been circumspect enough to think that in 
March of 1994 this would have become the problem that it has, cer- 
tainly we would have been more interested in doing so. 

But let me also say that we did have a plan of action as to what 
we were going to do, and that plan of action was to present a full 
report in a peer review journal with a technical report because 
there is so much data that, so much information that one single re- 
port — one single paper could not hold it. 

The strategy of the biostatistical center, biostatisticians was to 
do that, have a full technical report available to everybody and 
have a paper which would indicate what the findings were. 

Mr. Schaefer. If they asked for it. 



183 

Mr. Fisher. To publish the paper, pubUsh the paper in a journal, 
but a technical report on file. 

Mr. SCHAEFER. Now, let me also get this straight. You are saying 
that the fraud did not change results of the study; is this correct? 

Mr. Fisher. Yes, sir. 

Mr. Schaefer. I have a tough time figuring this one out. I am 
obviously not a medical doctor, but it seems to me if there was 
fraud and there was falsification of information, how could it not — 
how could it not? 

Mr. Fisher. Well, I think you raise a very important question 
and what you are asking is a very, very important one. And it gets 
into the problem of what do you do with the kind of data that were 
obtained, oh, falsified data or even ineligibility patients and all of 
that sort of thing. 

Now, there is a — the biostatistical community, the 
biostatisticians have a — their feeling about this, their belief, their — 
most of the major statisticians, Professor Pita from Oxford, who is 
one of the leading statisticians in the world, et cetera, et cetera, 
feel that — ^you have several choices when you are faced with infor- 
mation like this. You can either take all of the data out in doing 
your reanalysis, just throw it all away, or you can only take out 
that which was the flawed data. In other words, in the lumpectomy 
study of these thousands of patients, there were only six where it 
was demonstrated that there was truly fraudulent data. And you 
can get rid of those. 

So you can take them all out, take out just those that are fraudu- 
lent, or you can leave them all in. And this is known as the intent- 
to-treat principle. And that is — so that is a consideration. 

And the statisticians in the NSABP reached their own conclusion 
that they felt, and that was one of the things after the April — dur- 
ing the entire process, their intent was to use the intent-to-treat 
principle. 

Mr. Schaefer. All right. If we leave this information in, the 
fraudulent information in, do you not feel that the American public 
has a right to know? 

Mr. Fisher. We didn't leave it in, we took it out. We took it all 
out. 

Mr. Schaefer. Well, regardless, it just seems to me that women 
out there should know whether a study, conducted with American 
taxpayers money included fraudulent information. This was not di- 
vulged. That is what I don't understand and what the chairman 
doesn't understand and this whole committee doesn't understand. 
Why wasn't this divulged? 

Mr. Fisher. We had a plan for divulging it in a rational, regular 
basis. 

Mr. Schaefer. Yes. But that was after a number of women in 
this country had this surgery already. 

Mr. Fisher. Nothing from our findings indicated that they had 
any improper therapy. 

Mr. Schaefer. I guess — and I know my time is up and I will be 
back for some more questions, but it just seems to me for them to 
make an important decision like this on whether they go with 
lumpectomy or mastectomy, they have to have every possible shred 
of information that is out there. If there is fraudulent data in 



184 

there, they should be there for them to make this decision. Maybe 
only six of them turned up problems or whatever it is — this has to 
weigh heavily on whatever the final analysis is by this individual 
woman. That is what really concerns me. 

I yield to the chairman now and I will come back again. 

Mr. Brown. Thank you, Mr. Schaefer. 

Mr. Fisher. I would just like to, Mr. Schaefer, say and empha- 
size that these falsified data did not alter the data, and I would 
emphasize what I said in my preliminary remarks that women in 
this country do not need to be worried about having a lumpectomy, 
because not only have we, as I said, our data, but those from 
around the world and by other people, also indicate that 
lumpectomy is an appropriate therapy for breast cancer. 

Mr. Brown. Mr. Schaefer? 

Mr. Schaefer. I am sorry, Mr. Chairman, I have to come back 
on this one. Well, for you and I, we will never have to worry about 
whether we decide to have a mastectomy or a lumpectomy, but 

Mr. Fisher. Well, sir, I have two daughters and I have a wife 
and I have females around me, and I think all of us have to worry, 
male or female. 

Mr. Brown. Continue. 

Mr. Schaefer. Yes. We all have a right to know. 

Mr. Fisher, you said if your reanalysis had produced evidence 
that the conclusions of the study were affected by the data alter- 
ations, you would have reported those findings immediately. This 
seems to indicate that it is acceptable to not report findings of 
fraud unless the fraud is determined to change the findings of the 
study. Do you believe that? 

Mr. Fisher. No, I don't believe that necessarily. 

Mr. Schaefer. Is it yes or no? 

Mr. Fisher. During the — when the material was — we were really 
under an embargo until the ORI report 

Mr. Schaefer. I understand. I am asking you personally. 

Mr. Fisher. Yes. 

Mr. Schaefer. Is it acceptable to not report the findings, yes or 
no? 

Mr. Fisher. No, it is not acceptable to not report the findings. 

Mr. Schaefer. OK. Well then, the actions of somebody are cer- 
tainly inexcusable in this situation. 

Mr. Brown. Dr. Fisher, I am still unclear. My understanding is 
you — you said you removed the bad data, but my understanding is 
you published papers after that, you submitted papers after that 
with the bad data in them. Is that not correct? 

Mr. Fisher. Those data were included when the biostatistical 
group and so on decided that there was a need to invoke the intent- 
to-treat principle which meant leaving in all information. 

Now, there is another reason, too, for leaving in all of the St. Luc 
patients, which I failed to mention, and that is particularly would 
it — by taking out all of that data, let me repeat that, of the — there 
were 99 out of 1,500 patients in the St. Luc database which had 
falsified data. All of these other women were all of the women, and 
these falsified data related to patient entry information. 

But all the women were real patients, real women who all had 
breast cancer. They were all randomized. They were all followed 



185 

up. There was no alteration in the treatment that was given, and 
one of the things that by taking out all patients prevents you from 
doing is to determine the toxicities and so on of continued follow 
up. And that is one of the reasons that the biostatistical community 
feels very strongly about for leaving patients in. 

Mr. Brown. Didn't you just tell Mr. Schaefer that the bad data 
was taken out? 

Mr. Fisher. It was. In all of the 

Mr. Brown. Now you are saying some of it was included, you 
said in response to my question. 

Mr. Fisher. In January of 1993, we received a letter from the 
NCI stating that all of the data should be removed from what we 
interpreted to be any papers publishing new data, data that had 
never been published before. Up until that time, we had no guide- 
lines as to not to doing otherwise. 

Mr. Brown. But you had no guidelines not to do otherwise. But 
don't you know that if you are including this bad data that you at 
least have some disclosure of it without 

Mr. Fisher. I wish I had the biostatisticians here to address this 
because it really is an issue which is very strong in the statistical 
community which indicates strongly that there are ethical consider- 
ations and statistical considerations for leaving the data in. 

Mr. Brown. Well, maybe that is some theory that statisticians 
understand, but don't you disclose that information to the public 
when, statistics aside, if it is included, don't you disclose it and tell 
people so that the physicians know? 

Mr. Fisher. We couldn't do it with the embargo because it was 
not permitted to do that because we would have to admit where the 
data came from. By that time — that was the reason. 

Mr. Brown. You state in your testimony that you complied with 
NCI's request in January, that no data from St. Luc would be in- 
cluded in publications containing that new data. The subcommittee 
had a number of manuscripts submitted after that date where St. 
Luc data was included. Can you explain that discrepancy? 

Mr. Fisher. I will have to review that, because I don't know for 
sure which one. 

Mr. Brown. You were — my understanding is you were the author 
of this, and you cannot explain it, sir? 

Mr. Fisher. Which manuscript are you referring to? 

Mr. Brown. Well, the NCI request in January 1993, JNCI re- 
garding the St. Luc data. You are not aware of that? 

Mr. Fisher. January 

Mr. Brown. It was NCI's request in January of 1993 

Mr. Fisher. That was the letter that came to us, correct, telling 
us in subsequent publications. 

Mr. Brown. And then the paper was submitted a year later with 
the bad data. 

Mr. Fisher. Which paper? 

Mr. Brown. JNCI. 

Mr. Fisher. JNCI. Oh. Are you referring to the paper with 
endometrial cancer? 

Mr. Brown. Yes. 

Mr. Fisher. That paper. Well, that was a paper which was a 
more recent paper submitted to the JNCI which was the paper in- 



186 

dicating the information about endometrial cancer. And that infor- 
mation, in order to — that was providing information relative to the 
incidence of endometrial cancer and the deaths from endometrial 
cancer, and that was one of the most important papers I think we 
have ever done in that it was able, for the first time, to provide 
some quantification of the risks of getting endometrial cancer in 
breast cancer patients who received Tamoxifen, and it would, in 
order to be absolutely certain that we weren't underreporting 
deaths from endometrial cancer, we included the data from St. Luc 
Hospital because to have removed the data from St. Luc Hospital 
would have underestimated the number, could have underesti- 
mated the number of deaths from endometrial cancer, and that was 
what I was referring to previously about when you leave all pa- 
tients out, you cannot get toxicity or second cancers or this kind 
of thing. And that was why that was left and that was a very con- 
sidered decision. It was a decision that was considered — when I say 
"considered", I mean that it was given a lot of thought and the 
biostatisticians decided that was the appropriate course of action to 
take. 

Mr. Brown. So you have ably explained why you left the bad 
data in there. You have had five statisticians and X number of doc- 
tors that knew about this. Why didn't any of you notify the public? 
Why wasn't there any public disclosure other than these five stat- 
isticians and these doctors that knew about it? Why was there not 
any kind of disclosure? 

Mr. Fisher. There is in the paper. 

Mr. Brown. Not offered by you, but ordered by NCI, correct? Is 
it true. Dr. Fisher, that the paper was submitted without your dis- 
closing that information from St. Luc? 

Mr. Fisher. It was true what you say, but the final paper does 
have it. 

Mr. Brown. In late 1992, the informed consent form for preven- 
tion trial participates was amended to state affirmatively that no 
endometrial cancer deaths had been reported; is that correct? 

Mr. Fisher. In late 

Mr. Brown. In late 1992, that none had been reported due to 
Tamoxifen? 

Mr. Fisher. Yes. To consent form. 

Mr. Brown. By that time, there had been at least one cancer, 
uterine cancer death known to NSABP, correct? 

Mr. Fisher. No, sir. No, that is not so. At the time — at that time 
there were no incontrovertible cases of endometrial cancer known. 

Mr. Brown. You knew that there was a possibility of perhaps 
some link between Tamoxifen and the death, but it was not, your 
word, incontrovertible. So you maybe saw — did you see any link 
there at all or you just didn't see any absolute connection? 

Mr. Fisher. You are talking about one patient who had, and that 
I think the same patient was discussed previously by previous dis- 
cussers, and it was a case of a patient who died, a patient died and 
the death certificate of that patient was that this patient had 
sepsis due to — we don't need to get into the medical technicalities, 
although there was a — that this patient actually died of pulmonary 
embolus and this patient was coded and entered into the system 
of our system as having died of pulmonary embolus. 



187 

And the information was then obtained that this patient had an 
endometrial — had had an endometrial cancer, and this patient then 
had a breast cancer. She had an endometrial cancer, she had diver- 
ticulitis, she died of pulmonary embolus, and the question was, did 
she die from endometrial cancer or did she die with endometrial 
cancer? 

We did notify promptly Zeneca at the time of the regular annual 
report of second primaries of this endometrial cancer, would have 
been considered a second primary tumor, and they were informed 
that this patient had died. We reported that on 1-30-92. And then 
there was some question about whether the patient — ^you know, 
again I emphasize, and it has got to be emphasized and reempha- 
sized, that patients with breast cancer who subsequently get 
endometrial cancer and they die, do they die from the endometrial 
cancer, do they die from the breast cancer, or what? And it was not 
until later that we determined that there was most — the death was 
most likely due to endometrial cancer, but we were never sure. And 
even today, sir, we are not sure that this first patient died from 
endometrial cancer, because a recent review of autopsy slides by 
pathologists, several pathologists have suggested or raised the 
issue that this patient died, but she could have very well died from 
breast cancer period, because in the autopsy review they found 
cells in the bone marrow which were characteristic of breast cancer 
cells. 

And that — so that — we did call that patient a death from 
endometrial cancer in the report that you referred to about report- 
ing endometrial cancers, the paper, but that is — and that is what 
is called an endometrial cancer in that paper. But we did report 
this to ICI as a death from endometrial cancer to Zeneca on 12-13- 
93, when we felt that the autopsy report became available which 
said endometrial cancer, but there is still a question about that and 
about one other death from endometrial cancer where the same sit- 
uation prevailed. 

It is a very, very difficult problem for the pathologists, for the 
physician to determine whether you have — here you have a woman 
with breast cancer, she receives Tamoxifen for her breast cancer, 
and then subsequently she gets an endometrial cancer and subse- 
quently to that she dies. And to say that she dies specifically from 
endometrial cancer under those terms is very difficult. 

And if one goes back and looks at the literature of patients who 
have been thought to have died in several papers that have been 
reported, there is evidence that deaths in those papers reported 
were also related — were difficult to say were due to endometrial 
cancer and not the breast cancer. So this is — it is really truly a 
medical dilemma. 

I am not — I reemphasize firmly that this was not withheld in any 
way to conceal a death from endometrial cancer or two deaths from 
endometrial cancer; it was purely a judgment related to doctors, 
pathologists, getting data and so on. 

Mr. Brown. Well, I understand as well as a layperson can the 
complexities of the medical analyses and the difficulty of determin- 
ing what she might have died of if she had three or four things, 
any one of which could have caused her death. 



188 

How can you affirmatively say in a consent form that she did not 
die of endometrial cancer, then? How can you say that? 

Mr. Fisher. These — the original consent form had — what you are 
referring to said there were no deaths from endometrial cancer. 

Mr. Brown. Yes. 

Mr. Fisher. And that was a — that is an unfortunate thing that 
sentence got there. The original draft of that consent form did not 
have that statement, the earlier draft, and then that was there, 
and that consent form, that statement of no deaths from 
endometrial cancer — let me just — may I just — do you mind if I just 
talk a little bit about this, because this is a very difficult problem 
relative to the consent forms. 

I would only say that we were involved completely in the consent 
form formulations with the pharmaceutical manufacturers, with 
the NCI, and the NSABP, and the FPA, and others when these 
various consent forms were formulated. 

And as a matter of fact, the statement of no deaths from 
endometrial cancer put into that consent form, we believe, and 
have reason to believe, that statement, sentence was put in by the 
NCI in their review of the — in their — in the preparation of the con- 
sent form. 

Mr. Brown. Had you, Dr. Fisher, at the time of that death, when 
speaking with NCI in understanding what had been the consent in 
the consent form 

Mr. Fisher. Excuse me, sir. The fact that it said no death was 
absolutely right as far as we were concerned in that there were no 
specific deaths at that point in time which the NSA 

Mr. Brown. So you really didn't know that. If I could. Dr. Fisher, 
you really didn't know that was the case, you only knew that she 
could have died from any one of two or three or four causes; you 
didn't know — you couldn't say which one for sure, you couldn't say 
which one not for sure. 

Mr. Fisher. That is exactly right, and that is why we reported 
to the Zeneca on 1-30-92 that there was a patient who died, she 
died and she had that — so that patient was reported. We didn't put 
in a cause of death because we weren't sure of what the cause of 
death was. But she was a patient who had died of — actually, the 
second sheet, these patients were considered a second primary, peo- 
ple who got an endometrial cancer were second primary cancers 
and they were looked upon as like somebody who would develop a 
lung cancer or a colon cancer, and they weren't treated any dif- 
ferently. 

Mr. Brown. How many women might have signed up for that 
study having — having this statement that no one had died of uter- 
ine cancer? 

Mr. Fisher. I can't answer that, sir. But let me just 

Mr. Brown. Can you guess? Is it hundreds? Is it thousands? 

Mr. Fisher. I just don't know, because I don't know 

Mr. Brown. How long did the statement remain in there before 
it was corrected at the prodding of NCI, is my understanding? 

Mr. Fisher. We reported the fact — we reported this in October — 
end of October, October 31, to our group meeting in Chicago. The 
reports of that case and two other cases were reported, and that 
was at that time, incidentally, at that meeting, where it was re- 



189 

ported Zeneca personnel and NCI personnel were present, they 
were present, and this was reported at that time. 

But the consent form that was in use, the one you are referring 
to, I would only like to call your attention to the fact that in that 
consent form, it was vigorously described that there were all kinds 
of toxicities resulting from the use of Tamoxifen, and it was pointed 
out that Tamoxifen, women on Tamoxifen developed endometrial 
cancer in those three times as great, and we also showed that there 
were deaths from taking Tamoxifen due to blood clots and 
thromboembolism and other complications. So women going into 
this study knew that they were taking a drug that did have serious 
side effects. 

Mr. DiNGELL. Doctor, can you tell us how long this statement 
was in effect, that no one had died of endometrial cancer? 

Mr. Fisher. Subsequent to the meeting in Chicago which, as I 
say, was October 31, as I recall, there was — right after that we did 
meet with the NCI in November, at which time — November or De- 
cember, at which time the material, a more updated version was 
presented to the NCI officials, and a plan, an affirmative plan was 
immediately started to change the consent form. And this was im- 
plemented. And again, when we began to do that, make that 
change, again I would call your attention to the fact that changing 
a consent form is not — involves many different agencies. It involved 
the NCI, it involved the FDA, OPRR, it involved the physicians of 
the NSABP, and so on. So we had a whole long list of people or 
agencies who wanted to take part in the designing of the new con- 
sent form and they all had their own way — intention of what could 
be done. 

At the same time, Zeneca had called to our attention, as did 
other people, that there were other complications that needed to be 
put into that consent form, complications which related — things 
that had to be said in that consent form about pregnancy, about 
BES, about alopecia, about second cancers, second intestinal can- 
cers, and there was a continuous interplay between all of the peo- 
ple I mentioned, together with all of the people who were inter- 
ested in getting these various complications into the consent form. 
This led to continuous dialogue back and forth from our people in 
the NSABP with all of the government agencies, and virtual 
gridlock existed. 

Mr. DiNGELL. All right. Now, Doctor, the purpose of the informed 
consent form is to see to it that the person who participates in the 
test knows all of the risks to him or her that are associated with 
that participation; isn't that true? 

Mr. Fisher. Yes, sir. 

Mr. DiNGELL. That is the whole reason for it. So here, then, for 
a period of about a year, actually over a year, this form did not in- 
dicate the risk of endometrial cancer from the test; isn't that right? 
Nor did it indicate, nor did it indicate during that period of time 
that patients were at risk of cancer and possible death from the 
use of Tamoxifen in connection with this test for a period of better 
than a year; isn't that so? 

Mr. Fisher. No, sir. The consent form did indicate that there was 
a threefold risk of getting endometrial cancer. That was in the con- 
sent form. 



190 

Mr. DiNGELL. So the — it said the risk of cancer, but it did not say 
the risk of death of cancer. 

Mr. Fisher. No, sir. 

Mr. DiNGELL. And it may be that people still equate cancer with 
death, but it may be they don't. But if you just say you get cancer 
from this, it is not quite as true as saying people have died of can- 
cer from this. So for a period of a year, this form was not brought 
current with the real state of facts, was it? 

Mr. Fisher. It was several months I believe, I am not sure; I 
don't have the 

Mr. DiNGELL. All right. Now, can you name any of the agencies 
that you said had to sign off on this change who would have ob- 
jected to assuring the fullest possible information to the partici- 
pants in the test with regard to the real level of risk? 

Mr. Fisher. I am sure they would not have. 

Mr. DiNGELL. So it is not a realistic assumption that they would 
have resisted a change which would have better informed the peo- 
ple who were participating in the test, is it? 

Mr. Fisher. It wasn't a matter I guess of resisting; it was a mat- 
ter at the time of the back and forth that had to take place to do 
it. 

Mr. DiNGELL. Well, was this question ever laid before the agen- 
cies that you referred to? In other words, that you had the addi- 
tional — that you had deaths from endometrial cancer, was it laid 
before these agencies or was it not? 

Mr. Fisher. Well, certainly I think that it was laid before the 
agencies, yes. 

Mr. DiNGELL. Was it laid before them in connection with a re- 
quest to change the form? 

Mr. Fisher. Yes, sir. 

Mr. DiNGELL. Well, so we have come to the conclusion that for 
a little over a year that there was no public statement with regard 
to the risk of death of cancer which came out, nor was there a 
change in the form during that period. This is from the period late 
1992 to early 1994. 

How many women signed up for participating in this program 
during that period of time? 

Mr. Fisher. I have lost the thread of the date, sir, I just have 
lost it. 

Mr. DiNGELL. A goodly number, a few, none? 

Mr. Fisher. A goodly number, I would suspect. I am not sure 
how many. 

Mr. DiNGELL. When did you first start signing women up to this 
particular program? 

Mr. Fisher. The program opened on June 1, 1992. 

Mr. DiNGELL. So that is approximately commensurate with the 
date that you began to have the information with regard to the risk 
of death of endometrial cancer? 

Mr. Fisher. We had no deaths from endometrial cancer that 
were confirmed at that time, none. 

Mr. DiNGELL. How many have signed up now? 

Mr. Fisher. As of March, when the study was suspended, there 
were approximately 11,000. 



191 

Mr. DiNGELL. How many of them were warned with regard to the 
possible risk of death from endometrial cancer during this period? 

Mr. Fisher. I can't answer that. Their physicians were informed, 
and it was the job of the physicians to talk to the people about 
that. 

Mr. DiNGELL. As a matter of fact, what the people who signed 
up were told, according to the public pronouncements, was — and 
the form — that was no one had died of endometrial cancer as a re- 
sult of this test; is that right? 

Mr. Fisher. The consent form said no deaths, but the other thing 
that consent form — they knew when they were signing that consent 
form that there were a host of other complications, including 
endometrial cancer, and there were two deaths that had occurred 
from thromboembolism, so it wasn't 

Mr. DiNGELL. How many of these 11,000 women who signed up 
for this program were made aware of the two deaths? 

Mr. Fisher. These two deaths from thromboembolism? All of 
them; that was in the consent form. 

Mr. DiNGELL. All of them were made aware of the fact that there 
were two deaths? 

Mr. Fisher. Yes, sir. 

Mr. DiNGELL. That is somewhat at variance with your earlier 
testimony. 

Mr. Fisher. Not deaths with endometrial cancer, but deaths of 
pulmonary embolism or that sort of thing, the point being that all 
the women who signed up knew that tamoxifen was not a drug 
which had no side effects whatsoever. 

Mr. DiNGELL. But we do not quarrel about the fact that they 
were not told that there were no side effects to its use. We are dis- 
cussing very specifically the deaths which occurred here. How 
many were warned of those? 

Mr. Fisher. Well, at that time, when the prevention trials start- 
ed, we had no evidence as far as we can determine that there was 
specific — that there were patients who died specifically because of 
endometrial cancer. 

Mr. DiNGELL. Now, Doctor, the NSABP has provided the sub- 
committee with a number of documents in the last couple of 
months. One of these documents is a group of slides dated August 
1993. In these slides, there are at least two patients known to 
NSABP by August 1993 to have died from endometrial cancer, yet 
information on endometrial cancer deaths was not reported by you 
to NSABP meetings until late October of 1993, and was not re- 
ported to the drug manufacturer until December. 

Now, given the fact that the prevention trial was actively recruit- 
ing at this time, why was this information not immediately con- 
veyed in August, or earlier, so that the informed consent form 
could be changed? 

Mr. Fisher. Sir, to the best of my ability, I will try to explain 
that. 

The date on the slide was the date at which the Biostatistical 
Center closes the summary file. Now, what the summary file is, is 
that is the cutoff point that they use for preparing their informa- 
tion. I myself did not — when the person who made the slide, we put 
that on, and they usually do because that is the date that is set. 



192 

but I myself didn't get that information until the slide was being 
prepared for the meeting in October, so that is a discrepancy. 

It wasn't that I made that slide in August. The slide was not 
made in August; the slide was made in October. 

Mr. DiNGELL. Are you telling us that the NSABP didn't tell you 
about these events? Were you aware or not aware? 

Mr. Fisher. To the best of my knowledge, I was not aware of 
them. 

Mr. DiNGELL. OK, you were not aware. So then NSABP didn't 
tell you about these two deaths from endometrial cancer; shouldn't 
they have told you? 

Mr. Fisher. Well, again, the question about the two deaths from 
endometrial cancer, they should have told me, but there was still 
some question about whether these were deaths from or with 
endometrial cancer. 

Mr. DiNGELL. Well, you didn't know about it, so you didn't say, 
OK, we ought to look into this and fmd out. If these things had 
been brought to your attention, you would have said, we had better 
look into these things and find out whether these deaths are 
caused by endometrial cancer or something else; isn't that right? 
You certainly would have done that if that had been made avail- 
able to you. 

Mr. Fisher. I would have hoped we would have known sooner if 
there were deaths. 

Mr. DiNGELL. But they didn't tell you, and they didn't tell the 
manufacturer. 

Now, the manufacturer has got the possibility of lawsuits against 
him because of the fact that the people have died of cancer from 
taking this particular substance as part of a test. As a matter of 
fact, NSABP had the possibility of lawsuits against them. Univer- 
sity of Pittsburgh had the possibility of lawsuits against them. 

Nobody is notified about the fact that we have these cancer 
deaths flowing possibly — and I think we might even say "prob- 
ably" — from the use of tamoxifen. You were not made aware. Uni- 
versity of Pittsburgh was not made aware, others in NSABP were 
not made aware, the manufacturer is not made aware. Is this good 
administration? 

Mr. Fisher. We reported this to the group on October 31st to 
the 

Mr. DiNGELL. And the slide was made in August? 

Mr. Fisher. No, the slide wasn't made in August; the slide was 
made in October. The slide was made in October. 

Mr. DiNGELL. All right, the data was available in August. When 
was the slide made? 

Mr. Fisher. The slide was made for the meeting in October. 

Mr. DiNGELL. But the data was available in August then? 

Mr. Fisher. The data was — I don't know, I can't answer that. 

Mr. DiNGELL. It was significantly before the slide was prepared, 
because the preparation there 

Mr. Fisher. In preparing for the slide- 



Mr. DiNGELL. Because the data had been around for a while. 

Mr. Fisher. In preparing for this meeting, the August 30th data 
set, that was the cutofi" that they used. Now, whether that was 
known or not known, I don't know. 



193 

Mr. DiNGELL. Well, here is what the slide sheet on the slide says, 
it says endometrial cancer, parenthesis, EC, in B-14, as of August 
31, 1993. This means that prior to August 31 there was awareness 
in the test program that this was a problem. 

Mr. Fisher. I can't be sure. All I can say is that was when the 
summary file was closed in the data center — and I apologize for 
this, and I wish the biostatisticians who were responsible for this 
kind of thing were here; but they are not here, and I cannot answer 
the questions. 

Mr. DiNGELL. Let's go to the next question here. 

One argument that has been raised about audits, time limits and 
the flow of information, is that there is a lack of resources for that 
function. Is that a problem? 

Mr. Fisher. I am sorry? 

Mr. DiNGELL. Is there a lack of resources for dealing with ques- 
tions of audits, time limits, and information flow? In other words, 
do we not have the resources that is needed for that kind of thing? 

Mr. Fisher. There was a problem that existed. We — the peren- 
nial problem about funding in 1991, when our renewal grant appli- 
cation was submitted, we had asked for $180,000, which then we 
didn't get; we got $80,000. 

Mr. DiNGELL. The Chair is going to recognize the gentleman from 
Colorado, the gentleman from Colorado. 

Mr. Schaefer. Thank you, Mr. Chairman. 

I want to follow up on what the chairman just touched on. This 
member believes you are a very brilliant doctor, but I am concerned 
that there was a problem of underadministration. A lot of the infor- 
mation that was flowing out there through audits, et cetera, never 
seemed to get to you. So, therefore, it seems to me that we needed 
to add more administration. 

Now, did you ever ask NCI for additional funds for administra- 
tion? 

Mr. Fisher. Yes, we did, in the years 1989, 1990, 1991 and so 
on. We have in our records letters were sent because at that time 
our program was growing as it continued to grow, and there was 
either level or reduced funding at each year, and we actually had 
deficits in our budget, we requested support, and that sort of thing. 

Relative to personnel, the addition of personnel, we realized the 
need to do that. We actually recruited for quality assurance people, 
people to come in that had the kind of knowledge that could help 
us. We interviewed four people in the course of a short space of 
time, or over a period of time, and none of these people seemed to 
be the kind of people that we would want to have. It is very dif- 
ficult to get somebody in without any prior experience. 

There were other — there — so that those situations did exist. 

And let me also add one thing, and that is that Margie 
McLaughlin, who was the key auditor in the program for 20 years, 
resigned, retired, and that left the Biostatistical Center with a hole 
to replace. They had started replacing — training people and so on. 

Mr. Schaefer. All right. But did you specifically ask NCI for ad- 
ministration funds — not funds for the whole program, administra- 
tion funds? 

Mr. Fisher. I believe so. 



194 

Mr. SCHAEFER. You did ask Zeneca for funding. Did you ever ap- 
proach them on administration-only funding? 

Mr. Fisher. No, sir, never did. I would have thought that would 
have been an improper thing to do. 

Mr. SCHAEFER. Well, now I want to get into this reception thing 
for just a minute. So it is proper for them to pay for receptions, but 
not for administration funding. They indicated to us that they pro- 
vided the funds for the receptions. Now, who paid for the travel 
and lodging? 

Mr. Fisher. Travel to group meetings is included in the budget, 
travel for investigators to go to the group meetings. And I would 
emphasize that the Zeneca people did — that this is an entirely sci- 
entific meeting, the business of the group has no other function, 
and that meeting is totally for that. And as was explained, there 
was a reception at each of these meetings. 

Zeneca, at the beginning of this thing, sort of, you know, they 
more or less volunteered to do that years ago. Other pharma- 
ceutical companies made some contributions, and I personally 
never would go to Zeneca and ask them, will you give us money. 
We have people that work in my organization who have talked to 
them and so on. 

Mr. SCHAEFER. OK, getting back to the question. I want to un- 
derstand correctly. Zeneca and others provided funds for the recep- 
tions, but the travel to and from wherever it was, the lodging was 
built into your budget? 

Mr. Fisher. Most of the people who came to the meetings paid 
their own funds for travel, and that came — it could have or it could 
not have come from what they had been reimbursed for the patient 
payments, if they had some money left over, which I don't think 
they did; but that was the 

Mr. SCHAEFER. So it is fair to say, though, that in many in- 
stances it was taxpayers' funds that provided this? 

Mr. Fisher. Provided investigators to go to the meetings? 

Mr. SCHAEFER. Right, lodging and transportation. 

Mr. Fisher. Yes, sir. 

Mr. SCHAEFER. OK. The fraud on the data was discovered in 
1991; am I correct? 

Mr. Fisher. Yes, sir. 

Mr. SCHAEFER. The embargo was lifted when? 

Mr. Fisher. In April of 1993. 

Mr. Schaefer. ok. When you had these semiannual meetings, 
after April of 1993, were the people who came to these meetings 
made aware? 

Mr. Fisher. The plan was to present this, as we have said, in 
a final report; and actually that final report coincided with this 
current NSABP meeting. That was when it was planned to be 
done. At that time, the biostatisticians were to have a complete 
technical report which is in the progress report of this meeting, and 
that was when the whole thing was to be presented. 

Mr. Schaefer. OK. When was it presented? Ever? 

Mr. Fisher. At this meeting, right now. 

Mr. Schaefer. So it was actually done? 

Mr. Fisher. I presented some of this on Monday, Dr. Redmond 
and myself. 



195 

Mr. SCHAEFER. What Monday? When? What Monday? Just re- 
cently Monday? 

Mr. Fisher. Day before yesterday. 

Mr. SCHAEFER. So after the embargo was Ufted in April of 1993, 
you finally now presented the information that you had fraudulent 
data? 

Mr. FiSHER. Yes, sir. 

Mr. SCHAEFER. Why this long? 

Mr. Fisher. I guess it gets back to what was talked about before, 
and that has to do with when the issue of the plan was put in and 
also the other things that — where the results didn't make any dif- 
ference, and then along came a lot of other things which interfered 
with going ahead with this paper — at that particular time you were 
talking about, some of the consent form changes, we mentioned 
that; the problems with the prevention trial, there were people who 
were concerned about the prevention trial; the recruitment prob- 
lems; and there was always something that was coming in which 
interfered, which seemed to take precedence over presenting this. 

Mr. SCHAEFER. Well, maybe a lot of things in your mind did take 
precedence, but it would seem to me that this information probably 
should have been at the top of everybody's list, particularly once 
you tell me that the embargo has been lifted. When it was lifted, 
I don't see why you immediately could not have notified The New 
England Journal of Medicine or other comparable publications and 
gotten this word out as fast as possible. 

I might make one final comment here. I don't know what was 
spent, in taxpayers' dollars, in going to these semiannual meetings 
and I am sure such meetings are valuable and all that, but maybe 
some of those dollars could have been better used for administra- 
tion costs. 

Mr. FiSHER. Well, sir, let me, if I could, just for a moment, be- 
cause I think this is terribly important for — at least from my per- 
spective. 

When the NSABP started in 1970, when we began— 1971 in 
Pittsburgh — there were 24 members; and today, as you heard, 
there were over 500 members, and many thousands of physicians, 
over those years, which have participated. And one of the major 
goals that I had was to make this a community-based program, not 
limiting it to physicians. 

In the 1970's, there were precious few physicians at cancer cen- 
ters or major centers who participated in clinical trials, so I spent 
the next 10 years of my life, or more, trying to convince physicians 
in the community to participate in the clinical trial program; and 
I think that worked, and it did work, and community physicians — 
the quality of medicine through that mechanism has been up- 
graded. And so the scientific approach to practice of medicine has 
been improved. 

Now, those physicians came to these meetings. They didn't get — 
and as was made here, comments were made, whatever payments 
that were given to physicians for — not to physicians, but for the 
costs of putting patients into clinical trials did not meet the costs 
that were needed; and I think everybody involved would agree to 
that, that the cost for a patient in a clinical trial was $800 to 
$1,000, that sort of range. That money was to be used for a lifetime 



196 

of follow-up on those patients. It was a one-time payment, and 
therefore it took care of the costs of the administrative activity at 
each one of these community centers where they had to have a 
nurse, they had to have a data person, and it also paid for follow- 
ing up patients. 

This meant that private institutions all over this countrv and 
otherwise, other circumstances, were paying for these people, for 
these patients getting into the clinical trial. 

Now, we did have these meetings which, as I say, we had — first 
of all, the NCI used to — they wanted a meeting twice a year, and 
we complied for a long time, then it was reduced to once a year. 

As this grew and it became a 1,000 to 1,200 or 1,300 people 
would come to these meetings, we couldn't find places large enough 
to accommodate 1,200 or 1,300 people, and to keep places where 
the rates were low enough to be able to afford for all of these peo- 
ple to come; and we are talking about the lowest possible room 
rates and the lowest possible travel rates, et cetera. Aiid the reason 
for going around to different places in this country was based on 
geography — and you talked about Banff in Canada; yes, Canada 
has been a major contributor to the NSABP. At one time 40 per- 
cent of the patients in the NSABP came from Canada, from Van- 
couver to Montreal. Therefore we needed to have a place to accom- 
modate a thousand or so people, and it had to be in Canada; and 
we looked for those places and we found that at this particular 
place we could get a rate which was cheaper than in most other 
hotels. 

Mr. SCHAEFER. OK, but you made the decision then yourself on 
where to go — Hilton Head, S.C., or to San Francisco? 

Mr. Fisher. I think that was done, we had a meeting, our own 
group had a person who would, you know, find out where the 
places are, the rates they could get, all of this kind of thing, and 
that was how it was made. 

Mr. SCHAEFER. I would just suggest to the gentleman — I don't 
know, maybe I am really off base, but Hilton Head, S.C., I don't 
know if that is the cheapest place to go around or DisneyWorld or 
the Fairmont Hotel in San Francisco, unless you really pulled some 
strings. 

Mr. Fisher. Well, I would have to — I understand, but I would 
think that if I could — and maybe I could find out what the rate was 
that was paid. I have no idea. 

Mr. SCHAEFER. OK. 

Mr. Chairman, I think I am finished. 

Mr. DiNGELL. The Chair thanks the gentleman. 

Doctor, I have been looking at budget requests here. The total 
amount of your grant is about $7 million for the assortment of 
grants that you received; is that correct? 

Mr. Fisher. I believe so. Maybe — no, wait, I think that that is 
for the treatment trial. 

Mr. Dingell. Is that about right? 

Mr. Fisher. Treatment, yes. 

Mr. Dingell. Now, in your requests for audit, which is direct 
cost support for audit function of NSABP, and on 12-1-92 to 1-31- 
93, the original request was for $181,923, the review committee 
recommended $115,280; then the final NSABP request budget was 



197 

$84,295, so it was cut about 50 percent or thereabouts, down to 
$84,000. 

Now, I note that constitutes about 1 percent of the total amount 
of the expenditures; is that right? 

Mr. Fisher. I don't know. 

Mr. DiNGELL. If I understand correctly, that would be one audi- 
tor and a level of support which might include a modest but not 
excessive amount of travel or telephones or other support, and 
probably no clerical assistance or support. 

Do you regard that as being an adequate level of audit for a pro- 
gram of this magnitude? 

Mr. Fisher. No, sir. 

Mr. DiNGELL. Now, your final number, then, was $84,295. Now — 
so that was the final request that was made; that was not just — 
that was not what the government gave you, that was the final re- 
quest for budget support. 

Now, who was it that made that request? Was that you or was 
that somebody else? 

Mr. Fisher. Well, the $84,000 was not the final request, sir. 

Mr. DiNGELL. It says here on the paper that I am looking at that 
the original request, it says original NSABP requested direct cost, 
then review recommendation level, and then the next column says 
final NSABP request budget. So that is the request that was sub- 
mitted, $84,295; and of course you were given that, so you cut it 
that much before you ever submitted the request. 

Who was it that cut those moneys? Was it you or was it some- 
body else? Did you request it? Who requested it? 

Mr. Fisher. Any final fiscal sign-offs were done by me, and I am 
just not clear on this. 

Mr. DiNGELL. I am not criticizing the fact that you had a party 
at the affair that year that amounted to $84,000, but — well, actu- 
ally $80,000, but it does seem to me that you are more expansive 
with your expenditure for parties than you are with your auditing. 

Mr. Fisher. I don't believe it to be so, sir. I would hate to think 
that I would be so, after a lifetime of dedication to science, to have 
thought that I just find absolutely devastating, I really do. I don't 
think that I 

Mr. DiNGELL. Well, all I can say here. Doctor, is this, that ac- 
cording to the sheet which we have, which is a statement here that 
was submitted to the committee, it says the final NSABP request 
budget was $84,295, which happens to be exactly the amount you 
got, so that was the amount you requested. That is for an entire 
year of auditing, that is for dealing with things like compliance, ex- 
penditures of money, seeing to it whether appropriate notice is sent 
out with regard to changes in risk and things of that kind. 

Mr. Fisher. The peer review committee was the one that cut us 
down, the review recommended level. 

Mr. DiNGELL. That is interesting because the review committee 
recommended $115,280, but the final request was $84,000. 

Now, who cut it from $115,000 to $84,000? That is a cut of 
$30,000; that is a cut of about 25 percent. Now, who was it that 
cut the peer review committee's recommendation? Did you do it? 
Did somebody else do it? 

Mr. Fisher. I will have to let you know, sir. I can't give you 



198 

Mr. DiNGELL. Does this appear to have been a wise cut? 

Mr. Fisher. No. 

Mr. DiNGELL. Well, Dr. Fisher, the committee thanks you for 
your assistance to us. 

The Chair notes there is a vote on the Floor. I am going to run 
over and vote. I will be back in about 15 minutes, and the commit- 
tee will reconvene then in 15 minutes. 

[Brief recess. 1 

Mr. DiNGELL. The subcommittee will come to order. The Chair 
notes that our next witness is Samuel Broder M.D., Director, Na- 
tional Cancer Institute, National Institutes of Health. 

Dr. Broder, welcome to you. Doctor, you are aware of the fact 
that all testimony taken before this committee is taken under oath. 
Do you have any objection to testifying under oath? 

Mr. Broder. I understand. 

Mr. DiNGELL. Given that you are entitled to be advised by coun- 
sel during your appearance, do you desire to be advised by counsel? 

Mr. Broder. Certainly not. 

Mr. DiNGELL. You will note that copies of the rules of the com- 
mittee, the subcommittee, and the House are there to inform you 
of the limitations on the powers of the subcommittee as well as 
your rights as you appear here before the committee. 

If you have no objections, then, to testifying under oath, if you 
will please raise your right hand. 

[Witness sworn.] 

Mr. DiNGELL. Doctor, you may consider yourself under oath. Wel- 
come to the committee. 

TESTIMONY OF SAMUEL BRODER, DIRECTOR, NATIONAL 

CANCER INSTITUTE 

Mr. Broder. Good afternoon. Chairman Dingell and members of 
the subcommittee. I am Dr. Samuel Broder, Director of the Na- 
tional Cancer Institute. With the Chair's permission, I would like 
to submit my full written remarks for the record and briefly give 
an opening statement. 

Mr. DiNGELL. Without objection, the entirety of your statement 
will appear in the record. You are now recognized for such addi- 
tional comments or other comments as you choose. 

Mr. Broder. I will try to be brief. 

I am pleased to have this opportunity to bring you up to date on 
issues discussed at the April 13th hearing regarding the oversight 
and management of clinical research conducted by the National 
Surgical Adjuvant Breast and Bowel Project, referred to as NSABP, 
whose headquarters, as you know, is at the University of Pitts- 
burgh and also to bring you up to date on some of the corrective 
actions that are being taken to address the problems that this com- 
mittee has identified and that we ourselves have identified. 

Please allow me to state at the outset that progress has been 
made over the past few weeks in coming to an understanding with 
officials at the University of Pittsburgh regarding these very seri- 
ous issues. This is reassuring to us at the National Cancer Insti- 
tute, as it demonstrates the commitment of the University of Pitts- 
burgh to adhere to the highest ethical, moral and scientific stand- 



199 

ards. We particularly acknowledge Chancellor O'Connor and Senior 
Vice Chancellor Detre, as well as Dr. Ron Herberman. 

This commitment has led to certain areas of agreement that we 
believe will begin to reestablish the public trust in clinical trials in 
general and in the NSABP in particular. 

Specifically, I am very pleased to note the following: 

Accrual, or the necessary preludes to accrual for certain key clin- 
ical trials has resumed; 

New procedures are in place at NCI to deal with scientific fraud 
or misconduct, especially including more efficient notification to the 
public; 

The NSABP, as a cooperative group, will be recompeted through 
the usual peer review process; 

NSABP is in the process of selecting new scientific leadership 
and has developed a corrective action plan to address certain prob- 
lems in management and oversight. 

I will briefly address each of these points. 

Clinical trials serve as one of the foundation stones of the Na- 
tional Cancer Program. They are the real world test of which new 
therapies or preventive measures will show benefit for specific pa- 
tients. The components of all clinical trials include careful and sci- 
entifically appropriate protocol design, informed consent, data man- 
agement control, and publication of trial results. We must have all 
of these functioning together. 

The disclosures of data irregularities in NSABP trials and the 
lapses of oversight at NSABP headquarters in auditing and report- 
ing trials, including NCI's participation in these matters and in 
publishing reanalyses of the affected studies with clear disclosure 
to the reader, raise serious questions regarding the ability of 
NSABP to perform clinical research. In our system of cancer re- 
search, the primary responsibility for the conduct of clinical trials 
must rest with the grantee, but the National Cancer Institute has 
responsibilities as well. 

To do its job more effectively, NCI has established a new branch 
to oversee the quality assurance aspects of its clinical trials, and 
I can provide more details. We have provided more explicit guide- 
lines for auditing and for dealing with irregularities in clinical 
trials and NCI has undertaken a reevaluation of its quality assur- 
ance program and is moving to strengthen key comiponents. 

As for NSABP and its leadership, the group has submitted a cor- 
rective plan, including provisions for on-site oversight by NCI staff. 
We have found this plan for resuming certain clinical trials to be 
generally acceptable, subject to some clarification or modification 
and certainly subject to continuing oversight. Moreover, all parties, 
including the University of Pittsburgh, are committed to allowing 
the peer review process to determine the future location and sci- 
entific direction of its clinical trials administration. 

Two of our advisory boards, the Division of Cancer Prevention 
and Control Board of Scientific Counselors and the National Can- 
cer Advisory Board, both recommended in formal votes that accrual 
for new patients for the Breast Cancer Prevention Trial with 
tamoxifen be resumed as soon as possible. In addition, on June 7, 
1994, the Oncologic Drugs Advisory Committee of the Food and 
Drug Administration reviewed the recent information regarding 



200 

endometrial cancer deaths and other information and concluded the 
tamoxifen Breast Cancer Prevention Trial should continue, and 
also recommended that accrual be resumed. 

Unlike most treatment trials, the prevention trial requires risk 
assessment as the first step, but risk assessment does not mean an 
immediate assignment to either placebo or tamoxifen, as the proc- 
ess requires roughly one month's lead time. In general, resumption 
of accrual and risk assessments will be allowed at NCI-designated 
cancer centers, formal members of our Community Clinical Oncol- 
ogy Program — the so-called CCOP's, which are different from 
NSABP — and in the case of NSABP members, certain institutions 
with an acceptable audit record in the past 3 years. 

I will not review the entire history of the events that have led 
us to this point, as many issues have been addressed in previous 
testimony. I would like to bring to your attention several additional 
irregularities that have been identified at a few participating 
NSABP institutions. The problems range from incomplete record- 
keeping to deficient informed consent procedures. Following that, I 
would be grateful for permission to highlight a few of the main 
facts and steps that we have taken to strengthen some of our pro- 
cedures. 

At the last hearing, we informed you about the issue of serious 
data irregularities at St. Mary's Hospital in Montreal. The problem 
had been noted by NSABP in September of 1993, but was not 
brought to NCI's attention. In fact, NCI staff independently discov- 
ered the documentation regarding this problem in March of 1994 
during our site visit to NSABP headquarters. Currently, the Office 
of Research Integrity is still investigating this institution, but I be- 
lieve that there are data falsifications involved. 

Another of the institutions is the Memorial Cancer Research 
Foundation in Los Angeles, where NCI conducted an audit on April 
29th and 30th at the request of the local principal investigator, fol- 
lowing contact by a news reporter. Files for NSABP patients en- 
rolled on the B-06 study of breast-sparing surgery and those of 
other patients were reviewed by the NCI auditors. The major areas 
of concern noted by the NCI auditors included patient eligibility, 
the randomization process and the obtaining of informed consent. 

These findings confirmed the findings of the NSABP's own audit 
report of 1990 that a serious problem had been identified regarding 
the accuracy of the data reported at randomization. 

It is of note that neither the investigator in Los Angeles nor the 
National Cancer Institute had been notified of the concerns identi- 
fied in the 1990 NSABP audit report, nor had there been any fol- 
low-up action by NSABP to correct these problems. We have been 
informed that the NSABP report, in fact, languished in obscurity 
in an unmailed envelope in the NSABP files. Correspondence from 
NSABP provided to the auditors appeared to or actually did con- 
gratulate the principal investigator in Los Angeles on the status of 
follow-up cases. This is incomprehensible to us. 

This matter has been referred to the Office for Protection from 
Research Risks, OPRR, and the local Institutional Review Board, 
which we abbreviate as IRB, because of concerns about possible 
breaches of patient confidentiality and patients not having been 
provided, or not having been given an opportunity to provide a 



201 

truly informed consent; and to ORI, for issues related to the sci- 
entific conduct of the trial. 

Other problems include South Nassau Hospital on Long Island, 
where patients were enrolled despite questionable eligibility. NCI 
had been given verbal notification that the NSABP had suspended 
the institution in follow-up to the audit report. However, from what 
we can tell, this was never done and none of the additional follow- 
up reports were provided to us. 

I want to echo Chancellor O'Connor's comments that you heard 
earlier regarding the important contributions of this committee's 
staff in helping us identify and sort through these very complex is- 
sues. 

Other NSABP audit sites such as Tulane and Louisiana State 
University demonstrated certain data management deficiencies. 
NCI has conducted two audits at each of these sites. At the initial 
audits, which were arranged with essentially no advance notice, 
there were substantial portions of missing data on eligibility and 
other matters, perhaps due in part to the shortness of the notice; 
but in all measures, this was an unacceptable site visit situation. 

Both institutions have developed new data management plans 
and have taken corrective actions to ensure that these kinds of 
problems are minimized or don't occur in the future. In the case of 
Tulane, we have tentatively agreed on a conditional resumption of 
accrual based on their recruitment of a new leader for clinical 
trials, the development of a comprehensive data management sys- 
tem, and also provided that they expeditiously clear up outstanding 
delinquent data and residual issues from the May 1994 audit. They 
must also make adequate provisions to ensure the performance of 
their affiliates, and they must agree and have agreed to a reaudit 
in roughly 6 months. 

We hope that a similar plan can be worked out for Louisiana 
State University. 

The NSABP uses as its testing ground real surgeons, the kind of 
individual seen by the average American when they seek treatment 
for cancer. However, this characteristic, which is a virtue in many 
ways, can also pose special challenges for adhering to data manage- 
ment and reporting requirements. 

For our part, NCI requested changes in NSABP audit proce- 
dures, including on-site auditing, auditing more patient charts and 
providing to NCI fixed annual audit schedules long before these is- 
sues had become front page news. Unfortunately, NSABP did not 
comply, and NCI staff did not take action to compel compliance. 

Regrettably, our staff routinely accepted late reports from 
NSABP with their promissory notes to do better in the future, or 
with their explanation that the responsibility for conducting high- 
priority studies had strained their ability to issue timely reports, 
or sometimes with no explanation. This will not happen again. 

The NSABP probation period requires submission of an audit 
schedule for the coming month, the performance of on-site audits 
and the auditing of 10 percent of all charts. Timely reporting is re- 
quired; I believe we have their attention. An electronic system for 
alerting NCI staff to expect and to review site visit reports for all 
cooperative groups is being established. 



202 

I would like to very briefly turn to one other area. As of June 
10, 1994, 287 of 293 Institutional Review Boards have reviewed 
and approved the revised protocol and consent form, and approxi- 
mately 3,000 patients have signed the revised consent in connec- 
tion with the tamoxifen studies that we have discussed earlier. 
There is a "Dear Participant" letter that was targeted to all women 
and was required as of April of 1994; and we will explore other 
ways of communicating directly with the women in our studies. 

Finally, we have initiated recovery of funds in instances where 
scientific fraud or misconduct was identified. Recovery of funds in 
cases of a finding of scientific fraud or misconduct is now standard 
operating procedure at the National Cancer Institute and is one of 
the action items in our manual issuance on this topic. 

Also, the University of Pittsburgh has been instructed to provide 
criteria for ineligible or inevaluable patient accrual that would be 
subject to reimbursement under a condition of the grant that has 
been in place since April of 1993 — again, before these issues be- 
came newspaper items. 

Mr. Chairman, this has been a very difficult time for everyone 
involved in the events that have occurred over the past few 
months. There has been much criticism of all parties involved and 
much concern about the future. We have listened carefully to these 
comments and will do our best to address them. All of us, espe- 
cially those in the scientific community, must keep in mind that 
the uncertainties and challenges we face in these matters amount 
to nothing compared to what the average cancer patient must face. 

Cancer patients and their loved ones are our constituency. 

Thank you for the opportunity to appear before you today. 

[The prepared statement of Dr. Broder follows:] 



203 



Statement of Samuel Broder, M.D. 



Good morning, Chairman Dingell and members of the Subcommittee. I am Dr. Samuel 
Broder, Director of the National Cancer Institute. With me today are Dr. Bruce Chabner, 
Director of the Division of Cancer Treatment, Dr. Michaele Christian, Acting Chief of the 
Clinical Trials Monitoring Branch, DCT, and Dr. Leslie Ford, Acting Deputy Director of the 
Division of Cancer I*revention and Control, NCI. I am pleased to have this opportunity to 
bring you up to date on issues discussed at the April 13 hearing regarding the oversight and 
management of clinical research conducted by the National Surgical Adjuvant Breast and 
Bowel Project (NSABP), whose headquarters is at the University of Pittsburgh, and the 
corrective actions that are being taken to address these problems. 

Please allow me to state at the outset that progress has been made over the past few weeks in 
coming to an understanding with officials at the University of Pittsburgh regarding these very 
serious issues. This is most reassuring to us at NCI, as it demonstrates the commitment of 
the University of Pittsburgh to adhere to the highest ethical, moral, and scientific standards. 
We particularly acknowledge Chancellor O'Connor and Senior Vice-Chancellor Detre, as 
well as Dr. Ron Herberman. This commitment has led to certain areas of agreement that we 
believe will begin to re-establish the public trust in the clinical trials process of the NSABP. 

Specifically, I am pleased to note the following: 

1. Accrual for certain key clinical trials has resumed; 

2. New procedures are in place at NCI to deal with scientific fraud or misconduct; 

3. The NSABP, as a cooperative group, will be recompeted through the usual peer 
review process; 

4. NSABP is in the process of selecting new scientific leadership and has developed a 
corrective action plan to address problems in management and oversight. 

I will address each of these points in greater detail. 

Clinical trials serve as one of the foundation stones of the National Cancer Program. They 
are the real-world test of which new therapies (or preventive measures) will show benefit for 
specific patients. The key components of all clinical trials include careful and scientifically 
appropriate protocol design, informed consent, data management, quality control, and 
publication of trial results based on thorough statistical analysis. When all of these 
components are in place and functioning smoothly, we can have confidence in the results of 
clinical trials. When one or more of these components breaks down, we must identify the 
problem as quickly as possible and take action to repair it. The disclosures of data 
irregularities in NSABP trials and the lapses of oversight at NSABP headquarters in auditing 
and reporting trials, and in publishing reanalyses of the affected studies with clear disclosure 
to the reader, raised serious questions regarding the ability of NSABP to produce valid 
research. The primary responsibility for the conduct of clinical trials must rest with the 
grantee. But NCI has responsibilities as well. 



204 



To do its job more effectively, NCI has established " new Branch to oversee the quality 
assurance aspects of its large-scale clinical trials. More explicit guidelines for auditing and 
for dealing with data irregularities in clinical trials have been set in place, and NCI has 
undertaken a re-evaluation of its quality assurance program and is moving to strengthen key 
components. As for the NSABP and its leadership, the group has submitted a corrective plan 
of action, including provisions for on-site oversight by NCI staff. NCI has found this plan 
for resuming trials to be generally acceptable subject to some clarification or modification and 
continuing oversight. Moreover, all parties-including the University of Pittsburgh— are 
committed to allowing the peer review process to determine the future location and scientific 
direction of its clinical trials administration. 

Accrual has been opened at selected sites to certain key clinical trials. For selected treatment 
trials in breast cancer and rectal cancer, this means patients may be entered into trials 
immediately. Two of our advisory twards, the Division of Cancer Prevention and Control 
Board of Scientific Counselors and the National Cancer Advisory Board, both recommended 
in formal votes that accrual for new patients for the Breast Cancer Prevention Trial with 
tamoxifen be resumed as soon as possible. In addition, on June 7, 1994, the Oncologic 
Drugs Advisory Committee of the Food and Drug Administration reviewed the recent 
information regarding endometrial cancer deaths and other information and concluded that the 
tamoxifen Breast Cancer Prevention Trial should continue, and recommended that accrual be 
resumed. Unlike most treatment trials, the BCPT requires risk-assessment as the first step, 
but risk assessment does not mean an immediate assignment to receive either placebo or 
tamoxifen, as the process requires at least one month's lead time. In general, resumption of 
accrual and risk assessments will be allowed at NCI-designated Cancer Centers, formal 
members of the Community Clinical Oncology Program (CCOPs), and in the case of NSABP 
members, certain institutions with an acceptable audit record in the past three years. 

Currently NSABP is in the process of choosing new scientific leadership and has informed us 
that they hope to complete that process by September 1, 1994. As you know. Dr. Ron 
Herberman has been acting as the interim chairman of NSABP. Once the new leadership is 
in place, we hope to allow roughly one year to permit all potential grantee-competitors to 
prepare grant applications for components related to statistical support, auditing, etc., and 
submit them for peer review. 

With your permission, I will not review the entire history of the events that have led us to 
this point as many issues have been addressed in previous testimony; however, I would like 
to bring to your attention several additional irregularities that have been identified at a few 
participating NSABP institutions. The problems range from incomplete record-keeping to 
deficient informed consent procedures. Following that, I would like to highlight a few of the 
main facts and the steps we have taken to strengthen some of our procedures. 

At the last hearing, we informed you about the issue of serious data irregularities at St. 
Mary's Hospital in Montreal. The problem had been noted by NSABP in September 1993 
but was not brought to NCI's attention. In fact, NCI staff independently discovered the 



205 



documentation regarding this problem in March 1994 during our site visit to NSABP 
headquarters. Currently the Office of Research Integrity (ORI) is still investigating this 
incident. 

Another of the institutions is the Memorial Cancer Research Foundation in Los Angeles, 
California, where NCI conducted an audit on April 29-30, at the request of the local 
Principal Investigator following contact by a news reporter. An NSABP audit of this 
institution had been conducted and a report prepared on April 3, 1990; however, apparently 
neither the Principal Investigator in Los Angeles nor NCI had been provided with a copy of 
the report, which identified several inconsistencies and incomplete records. Files for NSABP 
patients enrolled on the B-06 study of breast-sparing surgery, and those of other patients, 
were reviewed by the NCI auditors. The major areas of concern noted by the NCI auditors 
included patient eligibility, the randomization process, and the obtaining of informed consent. 
These findings confirmed the findings of the NSABP's own audit report of 1990 that a 
serious problem had been identified regarding the accuracy of the data reported at 
randomization. 

Neither the investigator in Los Angeles nor NCI had been notified of the concerns identified 
in the 1990 NSABP audit report, nor had there been any follow-up action by NSABP to 
correct these problems. We have been informed that the NSABP report was attached to an 
unmailed envelope in the NSABP files. In fact, correspondence from NSABP provided to 
the auditors congratulated the Principal Investigator on the status of follow-up cases. This 
matter has been referred to the Office for Protection from Research Risks (OPRR) and the 
local Institutional Review Board (IRB) because of concerns about possible breaches of patient 
confidentiality and patients not having provided informed consent, and to ORI due to issues 
related to the scientific conduct of the trial. 

Other problems include South Nassau Hospital on Long Island where two patients were 
enrolled despite questionable ehgibility. NCI had been given verbal notification that NSABP 
had suspended the institution in follow-up to the audit report; however, this was apparently 
never done and none of the additional follow-up reports were provided to us. [Perhaps I 
could take this opportunity to acknowledge the diligence and effectiveness of your staff in 
helping us identify and sort through these issues.] 

Other NSABP audit sites such as Tulane and Louisiana State University demonstrated certain 
data management deficiencies. NCI has conducted two audits at each of these sites. At the 
initial audits, which were arranged with no advance notice, there were large amounts of 
missing data on eligibility and other matters due in part to unavailable charts. When we 
returned in May after more notice, the majority of charts and missing data had been located 
and were available for review. In addition, both institutions have developed new data 
management plans and taken corrective actions to ensure that these kinds of problems don't 
occur in the ftiture. We have tentatively agreed to a conditional resumption of accrual at 
Tulane, which developed a comprehensive data management system, providing that they: 
expeditiously clear up outstanding delinquent data and residual issues from the May 1994 



206 



audit; make adequate provisions to ensure the performance of their affiliates; and agree to a 
reaudit in six months. We hope that a similar plan can be worked out for LSU once we have 
received additional details regarding their corrective plan. 

One of the difficulties we faced was that NSABP, one of nine major clinical trials cooperative 
groups, had a philosophy of using a less stringent auditing system than the other groups. 
This was due to several factors, including the very large, community-based surgical 
membership of NSABP, its long history of productivity, its proud reputation, and its insistent 
adherence to its own traditions for auditing and quality control. The community-based nature 
of NSABP surgeon participants is one of the group's greatest strengths: Their results reflect 
the ability of community-based surgeons to apply new treatments and new surgical methods. 
The NSABP uses as its testing ground the real doctors who the average American will see 
when seeking treatment. However, this same characteristic also can pose special challenges 
for adhering to data management and reporting requirements. 

For our part, NCI requested changes in NSABP audit procedures including on-site auditing, 
auditing more patient charts, and providing to NCI an annual audit schedule. Unfortunately, 
NSABP did not comply and NCI staff did not take action to compel compliance. 
Regrettably, our staff routinely accepted late reports from NSABP with their explanation that 
the responsibility for conducting high priority studies had strained their ability to issue timely 
reports, or sometimes with no explanation. This will no longer happen. The NSABP 
probation period requires submission of an audit schedule for the coming month, the 
performance of on-site audits, and the auditing of ten percent of all charts. Timely reporting 
is required. An electronic system for alerting NCI staff to expect and to review site visit 
reports for all cooperative groups is being established and will be operational within 90 days. 
In addition, we have met with the Clinical Cooperative Group chairmen, who have agreed 
that there should be common standards for all groups in terms of what constitutes an 
acceptable audit. These criteria are being ''sveloped in conjunction with the groups and wiU 
be incorporated into the new NCI on-site monitoring guidelines. Adherence to these criteria 
is included in the terms of award for all cooperative group grantees. 1 should point out that 
these will represent minimum criteria, and individual Groups would certainly have the option 
to impose additional (more stringent) requirements for their member institutions. 

There will also be cooperative group-wide standards for auditing, reporting audit results, and 
dealing with instances of fraud or misconduct, as well as a group-wide requirement for ethics 
training and for credentialing new investigators. 

We have met with officials from the group and from the University of Pittsburgh to discuss 
how the remaining problems can be addressed. Several specific items have been discussed 
and agreements reached between NCI and the University of Pittsburgh: 

o We are informed that NSABP accepts the idea that its management needs improve- 
ment, and the Chancellor of the University of Pittsburgh has issued an apology for 
that institution's role in these problems; 



207 



o The University of Pittsburgh will resolve the inconsistency found as a result of the 
EMMES reanalysis of the B-06 (breast-sparing surgery) study between data and 
published journal articles regarding ipsilateral recurrence of breast cancer (i.e. 
recurrence in the same breast); 

o Each week, NCI staff will be on-site at NSABP headquarters in Pittsburgh to enhance 
cooperation efforts and to provide oversight. Pittsburgh will provide appropriate 
space; 

o NCI has raised concerns regarding remaining problematic issues, including a 1993 
memorandum from Dr. Carol Redmond to Dr. Bernard Fisher noting her serious 
concerns about protocol design and informed consents, and we expect a response from 
NSABP; and 

The University is proceeding with a broad inquiry into NSABP administrative issues, 
data irregularities, and human subjects protection in coordination with the Office of 
Research Integrity and the Office for Protection from Research Risks. 

The NCI is committed to working with the University of Pittsburgh and the new leadership of 
NSABP to ensure that research continues under careftil and appropriate scientific and 
administrative oversight. We believe that the leadership of the NSABP should be given an 
opportunity to heal any remsiining wounds, galvanize the membership, and develop and 
implement a scientific agenda. It is possible that there would be a physical relocation of the 
operations and statistical offices to new institutions and new sites. The operations offices of 
other cooperative groups have relocated in the past, and there are precedents for having the 
chair and the statistical center located at different sites. If this were to happen, the NCI on- 
site presence at the University of Pittsburgh would help expedite the process. But, once 
again, we are committed to allowing the principles of open competition and peer review to 
resolve these kinds of issues. 

We are still in the process of conducting a detailed audit of the B-06 (breast-sparing) study. 
Institutions that contributed 10 or more patients to the B-06 study, which accounts for 
approximately 1500 patients, will have been audited most likely by August. An NCI 
contractor will then conduct an independent statistical reanalysis. We estimate that this re- 
analysis will be completed by late summer or early fall, but I cannot absolutely promise this 
as a deadline. 

NCI staff are coordinating certain activities with the DHHS Office of the Inspector General. 
This will likely include reviews that will compare the research records at the NSABP 
statistical center with the computerized database used to analyze the study to ensure no errors 
or bias were introduced at the statistical center. 

1 would like to return briefly to the Breast Cancer Prevention Trial. One of the criticisms of 
this trial has been the lack of completeness and promptness of the informed consent process. 



208 



NCI has an obligation to provide accurate, up-to-date scientific information to patients who 
participate in our studies. We have taken some steps to quicken the process and we are 
exploring other new ways of communicating directly with patients to inform them of changes 
to informed consent. On April 18, as instructed by NCI, NSABP sent a notification to all 
participating sites that they must convey a "Dear Participant" letter to all BCPT participants 
by April 22, 1994. This step was intended to ensure that all women participating in the 
BCPT would have immediate notification regarding the information that had been distributed 
to the participating centers in January 1994. We felt this was necessary because the usual and 
customary process being followed, including review and approval by the local IRBs, is a 
lengthy one and many women had not been contacted directly. The information provided to 
women in the "Dear Participant" letter was the same as that included in the Zeneca "Dear 
Doctor" letter dated April 8, 1994. In addition to requiring direct notification to participants, 
we also are monitoring the reconsenting process very closely. Following instructions issued 
by the Office from Protection of Research Risks in November 1992, each IRB was provided 
with the full NSABP protocol and consent form, as well as the local institutions's version of 
the consent form. 

As of June 10, 1994, 287 out of 293 institutional IRBs have reviewed and approved the 
revised protocol and consent form, and 2,957 patients have signed the revised consent form. 
As I mentioned a few moments ago, however, aU women participating in the trial were 
targeted to receive notice of the changes as of April 22, 1994. 

We will continue to explore novel ways of communicating new information to patients 
quickly. One approach we are testing involves an electronic bulletin board that may be 
available in some doctors' offices or participating hospitals where patients could browse 
through information updates or even electronically indicate their "re-consent" after having 
read new information and discussed it with appropriate medical staff. We will explore many 
options to make more information available to patients more rapidly, and no single method is 
likely to serve all of our needs. 

Finally, we have initiated recovery of funds in instances where scientific fraud or misconduct 
was identified. Recovery of funds in cases of a finding of scientific fraud or misconduct is 
now standard operating procedure at the National Cancer Institute and is one of the action 
items in our manual issuance on this topic. Also, the University of Pittsburgh has been 
instructed to provide criteria for ineligible or inevaluable patient accrual that would be subject 
to reimbursement under a condition of the grant award that has been in place since April, 
1993. 

Mr. Chairman, this has been a very difficult time for everyone involved in the events that 
have occurred over the past few months. There has been much criticism of all parties 
involved and much concern about the future. We have listened carefully to these comments 
and will do our best to address them. In this process, our advisory groups have been most 
helpful. All of us, especially those in the scientific community, must keep in mind that the 
uncertainties and challenges we face in these matters, amount to nothing compared to what 
the average cancer patient must face. Cancer patients and their loved ones are our 
constituency. I believe that the result will be a stronger and more responsive clinical 
cooperative group program made up of dedicated, caring, principled scientists and physicians. 
We must expect nothing less. 

Thank you for this opportunity to appear before you today. 



209 

Mr. DiNGELL. Dr. Broder, the committee thanks you for your very 
helpful testimony. 

The Chair recognizes now my good friend from Colorado for such 
questions as he chooses. 

Mr. SCHAEFER. Thank you, Mr. Chairman. 

Dr. Broder, the tamoxifen trials have been resumed, as I under- 
stand it. Do you believe that NSABP is capable of monitoring these 
trials at the present time? 

Mr. Broder. Yes, I do. We will have oversight and we will have 
an individual up there a certain portion of the time, physically on 
site. They have provided us or have assured us that there will be 
space, and we believe that they — based on a number of things that 
they presented to us privately and in some of the testimony that 
you have heard — have a sincere commitment to this issue, and I 
believe that they have a can-do spirit and they will comply with us 
and comply with the needs of the public and with the interests of 
the Congress. 

Mr. SCHAEFER. I am sure that this committee, as well as a lot 
of the people in this country, agree that they have a new spirit, 
that we aren't going to get into the same problems all over again. 
What type of person are you going to have up there? Are you going 
to have a physician there or a doctor? 

Mr. Broder. We will try to have a professional full-time em- 
ployee of the National Cancer Institute up there a certain amount 
of time. I can't tell you the individual will be there 7 days a week, 
24 hours a day, but there will be an individual up there, and we 
will probably have a certain schedule or rotation since we, at the 
current time, cannot detail a single individual. 

We will probably establish a rotation to have an NCI professional 
on site periodically to oversee and to interact and to make sure 
that we are in full compliance with all of the issues, and that we 
are correcting the problems that have been identified. 

Mr. SCHAEFER. In your opinion, does NSABP currently have ca- 
pabilities to conduct their audits and generate audit reports that 
change that much that they can do that now? 

Mr. Broder. Thank you for the question, Mr. Schaefer. 

I want to make a point very briefly that while there are going 
to be some exceptions to what I am going to say, many of the issues 
that we have dealt with today were not necessarily the fault of the 
actual hands-on people doing the audits. Miss McLaughlin, for ex- 
ample, was a career employee doing audits. In my opinion, at least 
from what I can tell — and I would be prepared to be informed oth- 
erwise if I am wrong — basically she told it like it was when she en- 
countered problems; though we certainly should improve the audit 
process and the scheduling and so on. 

But the actual hands-on performance of the audits is not so 
much of the problem as what was done with the information, once 
gathered. I believe this is a recurring theme, that perhaps most of 
the staff that worked with me on this matter would agree, that 
there is an identification of a problem, it is sitting there, it stays 
in a file, and there appears to be no action and no apparent expla- 
nation for the inaction. So I believe that it is a matter of philoso- 
phy as much as it is resources and the commitment. You have to 
want to act on your information. 



210 

Mr. SCHAEFER. Right. I am sure you heard testimony just prior 
to this as regards these audit reports. Too many times, as I cer- 
tainly understand it, they didn't reach the point where they should 
have. So all of this information sitting out there as to all of these 
various problems that we have, and it never reached the proper 
people. 

Mr. Broder. Either it didn't reach the proper people or for what- 
ever reason, perhaps well justified or not, appropriate decision- 
making was not based on the facts as they were available. 

Mr. SCHAEFER. Do you believe that they are understaffed as far 
as the administration level goes? 

Mr. Broder. I do not believe that they are understaffed, and I 
am not sympathetic, at least excessively sympathetic to the argu- 
ment that most of the problems that we have heard today are fo- 
cused on resources. 

Yes, if you were at an appropriations hearing, I would certainly 
give you a very good defense of why we need a budget and why we 
need certainly the President's budget. But I don't believe that most 
of these issues that we have encountered today are focused on re- 
sources per se, and I actually feel that is taking us a little bit away 
from the point. 

Mr. SCHAEFER. So just with a little bit of streamlining in the 
present system, it could probably do the job without 

Mr. Broder. Well, I don't want to say that resources aren't rel- 
evant; they are always relevant, as we can certainly discuss, and 
I will also get into the issue of resource allocation should you wish 
me to, but I think that it is basically a philosophy, which also has 
to be a part of the process. 

We have to, without fear of favor, as much as human beings can, 
accept data as thev come in. That is what a scientist really must 
do, and auditing from my point of view is a specialized kind of 
science. You have to take the facts as they come in, not as you 
want them. It is very dangerous for a scientist or a doctor to ignore 
facts, and I think that in this situation, the recognition of the facts 
as they came in, and an appropriate decision-making process after 
those facts were known would have gone a long way to obviate 
many of the problems that we have. 

Mr. SCHAEFER. Well, at our April hearing. Dr. Friedman testified 
that two people had been added, assigned to the monitoring of the 
reports that came in from the University of Pittsburgh, and he 
stated he thought this was inadequate. Is that your opinion or can 
you even make an opinion? 

Mr. Broder. Are you asking whether we at NCI should be allo- 
cating more resources? 

Mr. SCHAEFER. No. Well, I am just saying, he stated that two 
people added is inadequate, it is not sufficient. 

Mr. Broder. I agree with that, but in agreeing with that I want 
to acknowledge that many of the problems which this committee 
has identified and that we have discussed or that NCI itself have 
identified are not linked to a resource issue. 

Yes, resources are necessary, and I pride myself in being pretty 
good in making a case for more resources, but I don't believe that 
is an issue before us today. Many of the issues that were identified 
earlier had to do with not acting on information as it came in, in- 



211 

formation that auditors had picked up, information that was in a 
file — memos that one investigator at NSABP would write to an- 
other investigator without a follow up. I believe that there were a 
lot of decision-making issues. 

We heard earlier today discussions about endometrial cancer 
deaths. I don't believe resources are the issue that we were talking 
about. 

Mr. SCHAEFER. OK. You also probably heard me — I referred, and 
the chairman did too, to this embargo thing in a couple of cases 
which prevented NSABP from announcing the presence of fraud in 
studies. Do you know under what authority such an embargo can 
be initiated? 

Mr. Broder. I would answer that in two ways. First, the embar- 
go is a request of a sister agency of the Public Health Service to 
not interfere with an ongoing investigation, and I understand the 
responsibilities that OKI has in this matter, but I believe you and 
others on the committee very carefully pointed out that the embar- 
go ended approximately in April of 1993, and well prior to the end- 
ing of the embargo, NCI and ORI staff had communicated to the 
NSABP that we expected an analysis published that would exclude 
the Poisson data and provide a notice to the reader. 

What we are talking about is an ethics issue, not a statistical 
issue. It is a notification to the reader; it is a disclosure to the 
reader that there is a problem. 

We had received assurances that the process was working and 
that a paper was being prepared. It was a surprise to us to learn 
that in fact, did not occur, and, please, I might want to add one 
other thing. 

There was a publication that came out in June of 1993 which ex- 
cluded the data from the St. Luc's site. There was a galley proof 
still available in April of 1993. The individual patients from St. 
Luc's Hospital appear to have been removed without a disclosure 
to the reader. 

So those are just the facts as I know them, and I have had dis- 
cussions with the editor of the New England Journal of Medicine 
on this topic. So it is certainly possible to argue, as you have suc- 
cessfully done, that there were substantial delays here, and we are 
sorry for them. 

Mr. ScHAEFER. It gets back to the people's right to know, as we 
indicated. 

One final question, Mr. Chairman. Was NCI responsible for al- 
tering the consent form as Dr. Fisher alleges? 

Mr. Broder. The NCI submitted an informed consent in re- 
sponse to a request by the Food and Drug Administration to spe- 
cifically address the mortality implications of endometrial cancer, 
and relied on Dr. Fisher's and others' presentation of information 
from their database. They were asked were there endometrial can- 
cer deaths and we were told no. We must rely on the grantee in 
these matters. I believe it was a joint decision. 

Mr. Schaefer. So was it only because Mr. Fisher had not re- 
ported the deaths? 

Mr. Broder. Well, as you heard his testimony today, he agreed 
with the inclusion of the statement, and from his point of view, it 
was valid as of the time of the insertion. Had we known then what 



212 

we know now, that statement would not be in there, and I submit 
to you that the grantee has a duty to inform us. 

One of the things that I noticed in the testimony up until now 
is that there was a great deal of discussion about reporting to 
Zeneca, the pharmaceutical company, and so on and so forth. As 
they said in Cool Hand Luke: What we have here is a failure to 
communicate. And basically I think that we should have known 
that there were deaths. We are the grantor, and if there are dif- 
ficulties of sorting them out, then we will be happy to participate. 

I don't think it is appropriate for us to learn at a scientific meet- 
ing as the company has expressed, that there is a new and unre- 
ported side effect. That puts us in the status of just another partici- 
pant in the study, and I reject that concept. We are the grantor. 

Mr. SCHAEFER. So you didn't know what Dr. Fisher knew. 

Mr. Broder. We learned at the public presentation, which was 
at the end of October. 

Mr. SCHAEFER. Yes. I couldn't agree with you more on that last 
one. 

Mr. DiNGELL. The Chair thanks the gentleman. 

Dr. Broder, the Chair would like to thank you for your very help- 
ful testimony to us today. We always are appreciative of your kind- 
ness and the very fine job you are aoing at NCI. 

You appeared before the subcommittee approximately 2 months 
ago and were aware of the problems in the NSABP trial sites at 
St. Luc's and were beginning to learn about the problems at St. 
Mary's. Based on your testimony today, it appears that NCI has 
learned a lot more about the management and operations at 
NSABP and its sites since that time. 

Can you characterize for the subcommittee what you found re- 
garding the management and operation at NSABP and its sites, 
please? 

Mr. Broder. I think on balance, we have observed that we must 
have a more organized and objective set of criteria for who can be 
a participating member; we must have more effective follow up and 
communication with the participants; we must have a process in 
which problems are identified quickly, and basically, if possible, 
corrected by the grantee, not simply corrected by the NCI coming 
in on an emergency basis. The final responsibility for these issues 
must lay with the grantee. 

I think that those are the major issues, that basically we must 
have a more organized and effective method of adhering to protocol 
eligibility and to following the clinical trials as they are designed. 
And I think we are working in that direction. 

I think Dr. Herberman has expressed an extreme commitment to 
this point, and I am rather pleased with the direction that he has 
been able to do an interim principal investigator. 

Mr. DiNGELL. Now, Doctor, we have heard a number of expla- 
nations as to why the problems have occurred. For example, lack 
of audit resources, and it does appear that there is a deficiency 
here; lack of an executive director to handle management adminis- 
tration; that the trials grew too quickly. I would like to get your 
thoughts with regard to the matters before the committee today. 

First, with regard to the issue of lack of audit resources, did NCI 
underfund audit requests made by Pittsburgh? 



213 

Mr. Broder. The short answer is no. I will elaborate if the Chair 
asks, but the answer is no, and I will provide you with the informa- 
tion either now or for the record. 

[The following information was received:] 



214 



f ^^ DEPABTMENT OF HEALTH & HUMAN SERVICES Public Health Sorvice 




National Institutes of Health 
_,__,_„ . -^T »^ TT ••«• National Cancer Institute 

MEMORANDUM BethesOa, Maryland 20892 



DATEi June 14, 1994 

FROM: Section Chief, GAB, OAM, NCI 

SUBJECT: NSABP Audit Support 

TOi Chief, GAB, OAM, NCI 



The following represents a comparison of the funds requested, 
recommended and awarded on CA12027-20, 21, 22 and 23 and CA37377- 
07, 08 and 09 for NSABP staff on-site audits of the NSABP 
participant institutions. It also includes a comparison of the 
amounts awarded and expended for CA12027-21, 22 and 23 and 
CA37377-09 for audit travel. (To date, we have not requested 
expenditures for years 07 and 08 on CA37377.) I have indicated 
the FY involved and the budget period. 

Attachment 1 compares the direct cost support requested by NSABP 
for the audit functions to the amount recommended by peer review, 
the amount subsequently requested by NSABP in response to NCI 
funding plans or Type 5 committed levels, and the amount awarded 
by NCI for audit functions for CA12027-20, 21, 22 and 23 and 
CA37377-07, 08 and 09. 

In Attachment 2, I have provided a comparison of the funds 
awarded for audit travel to the reported NSABP expenditures of 
travel funds for years 21, 22 and 23 of CA12027 and year 09 of 
CA37377. This is the information requested of Mr. Crouch in our 
April 8 and June 2 letters. The reason for reflecting only the 
travel amounts is that for the 21 and 22 years of CA12027, 
figures were only provided for the expenditures on travel. 
However, in the report on the 09 year of CA37377, he indicates 
budgeted and expended funds for auditors. In our review of the 
file, we can not find reference to a request for salaries 
specifically for auditors, so I have removed those amounts from 
the reported expenditures to allow comparability of cost pools. 
(The budgets for CA12027 included salaries and other costs and, 
therefore, those figures are included in Attachment 1.) 

I have attached copies of the information that Michael Crouch 
provided in response to our requests of April 8 and June 2, 1994, 
in which we asked for a description of the audits which were 
planned for each year as compared to the audits which were 
actually performed and the associated cost of each audit. 

If additional information is required, please let me know. 



215 



ATTACHMEMT 1 
DIRECT COST 80PPORT FOR ADDIT FUMCTION OF NSABP 



PROJECT/YEAR 

CA12027-20 

FY 91 
(2/1/91-1/31/92) 

CA12027-21 

FY92 
(2/1/92-1/31/93) 

CA12027-22 

FY93 
(2/1/93-1/31/94) 

CA12027-23 
FY94 
(2/1/94-1/31/95) 



ORIO. NSABP 
REQUESTED 
DIRECT COST 

S 52,076 



181,923 



REVIEW 
RECOMM. 

LEVEL 



FINAL 
NSABP REQ. 
BUDGET 



Unavail. 



115,280 



S 76,413 



84,295 



NCI 

AWARDED 
DIRECT COST 

$ 71,838 



84,295 



192,095 



202,851 



41,459 



CA37377-07* 

FY91 
(8/1/91-5/31/92 Ten months) 



Unspecified 90,100 



Unspecified 185,960 



Unspecified 21,555 



CA37377-08* 

FY92 
(6/1/92-5/31/93) 

CA37377-09* 

FY93 
(6/1/93-8/31/94)** 
TOTAL 



Unspecified Unspecified 49,759 



Unspecified Unspecified 92,923 



$601,005 



84,466 



174,803 



21,555 



45,136 



90,750 



$572,843 



* Requested support for travel only. 
** 09 year extended at no-cost for 3 months from 5/31/94 to 8/31/94. 

Definition of Columns 

Column 1 - Original requested levels found in application. If no detail was 
provided in the competing application for specific amounts requested for audit 
support in future years, then unspecified is indicated. 

Column 2 - Correct recommended level based on peer review summary statement. 
If no detail was provided for audit in future years summary statement 
description, unspecified is indicated. 

Column 3 - NSABP 's final budget request based on NCI funding availability in 
competing years (CA12027-21 and CA37377-07) or Type 5 commitment level shown 
on award notice. 



Column 4 - NCI's awarded amount based on funding plan level for each year and 
programmatic review of request. 



216 



ATTACHMENT 2 
COMPARISON OF AUDIT TRAVEL FUKD8 AWARDED TO EXPENDED 





NCI 


NSABP 




AWARDED 


EXPENDED 


PROJECT /YEAR 


TRAVEL 


TRAVEL 


CA12027-21 


$ 29,389 


$21,579. 


FY92 






(2/1/92 - 1/31/93) 







CA12027-22 26,713 

FY93 
(2/1/93 - 1/31/94) 

CA12027-23 86,851 

FY93 
(2/1/94 - 4/30/94) 

CA37377-09 83,393 

FY93 
(6/1/93 - 4/30/94) 



DIFFERENCE 



10,035.22 • 16,677.78 



3,683.44 83,167.56 



64,555.28 18,837.72 



TOTAL TRAVEL 



$226,346 $99,853.55 $126,492.45 



217 

Mr. DiNGELL. Now, then there is the next question of did the 
trials grow too fast? I don't believe that NCI was ever told that 
NSABP officials were concerned about the rate of recruitment and 
worried about the effect that this would have on the quality of the 
work being done; is that correct? 

Mr. Broder. Well, it is very difficult for me to understand that 
the trials were going too fast. There was a grant put in. The grant 
was put in by the NSABP to perform a trial. 

Presumably, the grantee knows what the grantee is asking in the 
trial, and that is a point that is a little difficult for me to follow. 
The trial was to do a major prevention study, and it had expecta- 
tions and growth. 

With the Chair's indulgence, I cannot resist getting back to the 
budget issue. I apologize. I need to define one point. 

At the type two or recompetition phase of a grant, an applicant 
will come in with a request. I am giving you numbers that are ap- 
proximate, but they are in the rough ball park. For example, one 
of NSABP's grants in the prior year, or the last year in the pre- 
vious cycle might have been $4.5 million, approximately. As they 
came in for the renewal or recompetition, they may request from 
us $9 million, or a 100 percent increase. The peer reviewers may 
cut that to $8.1 million, and since we do not have infinite growth 
potential we may end up finally giving the NSABP on that grant 
$6.5 million. Now, $6.5 million is a substantial level of growth com- 
pared to $4.5 million, but it is certainly a cut compared to $9 mil- 
lion. 

Mr. DiNGELL. It is about a 50 percent increase. 

Mr. Broder. Right. So it depends on how you want to look at it. 

The final programming of resources within the total award in all 
of these matters is in the hands of the grantee. The final decision 
is made when we receive a final NSABP budget request. That is 
what the grantee is telling us they need, and that was the point 
you were raising earlier. That is what the grantee is telling us, 
that they want to do — that is their best determination — and we ne- 
gotiate. We would always give more flexibility if the grantee asked, 
and sometimes there is money left over for procedural reasons at 
the end of a year and they would be most welcome to make a 
rebudgeting request. 

In addition, you should be aware that at the front end of the 
original request it is not uncommon for the auditing function to be 
unspecified, and the specification of the amount actually doesn't 
come in until the final budget negotiations. So I am sorry to have 
taken your time with the arcana of NIH grants management, but 
I certainly do not accept the principle that we at NCI inappropri- 
ately cut a deserving audit function. I think there was certainly 
less money to go across the board for all of the research effort, but 
the exact proportionality that one chooses to commit for budgeting 
for auditing is in the hands of the grantee. And, finally, as I say, 
many of the issues that we encountered don't have anything to do 
with resources. 

Mr. DiNGELL. Now, there the two other things that we have 
heard in the questions of the matter before us. One is the request 
of need for some kind of better management, for example, a direc- 
tor type position that could handle management administration. 



218 3 9999 05982 385 4 

Did the people at NSABP ever indicate that they would need 
some kind of assistance of this sort to handle the trials because 
they lack the capability to administer the project properly? 

Mr. Broder. No. And Dr. Fisher is a great man and has made 
great contributions. Anyone who knows NSABP will know that Dr. 
Fisher is the person who runs NSABP. 

Mr. DiNGELL. I got sort of the opinion that sometimes maybe it 
was somebody else running it; I was not altogether clear who was 
running it at this point. 

Mr. Broder. That is the basis of my statement. 

Mr. DiNGELL. Now, you heard Dr. Detre earlier this morning 
about the culture deference given the senior scientists. I think that 
deference is required, but I am curious, is the deference going be- 
yond their ability to address the problems that they confront in 
terms of actually administrating a project of this kind? 

Mr. Broder. I think you are raising an excellent point. Inherent 
in your question is, ate the attributes that require creative sci- 
entific thinking the very same attributes that are necessary for an 
executive of clinical trials administration? I think that is a valid 
point and we are certainly considering it. 

I think we may need to look at and pay more attention to the 
difference between the scientific and creative process, the genius 
which Dr. Fisher has, and he is a genius, in the development of 
certain aspects of the scientific outline of a study compared to how 
one executes that study and how one then pays attention to the in- 
credibly important details that are necessary to make the thing 
work. And I think that we will certainly look at that issue. 

Human nature is difficult to change, and of course if you are in 
charge of a study, sometimes you want to be in charge of every- 
thing, and that is one of the issues we will have to grapple with. 
But we are certainly looking into whether we should be more effec- 
tive at circumscribing activities, so that an administrator or admin- 
istratively oriented person may run the audits and possibly be in 
charge of certain issues related to quality assurance, and that need 
not be the same person that develops new scientific ideas, and 
often wouldn't be. There is no harm or shame in saying that. There 
is specialization in all fields. 

Mr. DiNGELL. Thank you, Doctor. Now, Doctor, the last time you 
were here we discussed the four endometrial cancer deaths in re- 
cent treatment trials. I note that there has now been a fifth that 
has been brought to the attention of NCI through notification. Can 
you tell us briefly about this particular death and the cir- 
cumstances that attend it? 

Mr. Broder. We were made aware of an additional death, I be- 
lieve that came to our attention in May of this year, that appar- 
ently had occurred in December of 1993. I would be happy to pro- 
vide the details for you for the record. I don't know the specifics 
of how the case was diagnosed. 

Mr. DiNGELL. Well, it appears that the NSABP learned of it in 
April, as a matter of fact April 8 of 1994, and they informed the 
NCI of it the first week in June. Is that right? 

Mr. Broder. I thought it was May, but it — I have staff here who 
could answer that question. 

Mr. DiNGELL. You would know better than I would. 



219 

Mr. Broder. I apologize. It was June 3rd. You are quite correct. 
June 3rd. 

Mr. DiNGELL. Doctor, it is always a privilege to have you here 
before the committee. I want to express my thanks to you, my com- 
mendations of the fine job you are doing. 

I would like to have a few concluding remarks. First to you. Dr. 
Broder, congratulations, and both for your testimony and for your 
leadership in this area. We think that it is important that you con- 
tinue to do so, because it is not the business of this committee to 
try and be the policeman of science; we want the scientists to do 
that themselves rather than us. 

To Doctors O'Connor, Herberman, and Detre, I want to commend 
you for your testimony today, gentlemen. I know it was not an easy 
day for you, and it is not the practice of this committee to make 
our days easy for our witnesses. But your testimony indicates that 
you have a solid understanding of the portions that contribute to 
the misfortunes in the NSABP matter, as well as remedial actions 
that need to be taken to make this institution a responsible, pro- 
ductive enterprise once again, and I commend you for that. 

We will observe the continued developments in this matter, in 
particular the implementations of the promised reforms with great 
interest, and we express to you our good wishes that you be suc- 
cessful in that, because it is important. 

You come from a great institution, one of which you are justifi- 
ably proud, and I know you want to make it better, and it is the 
good wishes of this committee that you shall be successful in mak- 
ing it better. 

We hope that this whole matter will now be brought to a happy 
conclusion, because we know the importance of the work that is 
being done here, and to the surprise of some in the scientific com- 
munity, this committee, this subcommittee, and this particular 
chairman have spent a great deal of time and effort and money to 
see to it that NIH is a success, it is adequately funded, that you 
scientists and educators and research specialists of different kinds 
are successful in your understanding because they are very impor- 
tant to us and to the country. 

So we express our thanks to all who assisted us today. We com- 
mend our witnesses; we give them our good wishes for continued 
success, and we call the committee to an adjournment. 

[Whereupon, at 4:03 p.m., the hearing was adjourned.] 

o 



84-510 (224) 



ISBN 0-16-046276-2 



9 780 



60"462764 



90000